Sampling networks of ecological interactions

Spatially explicit predictions of food web structure from regional-level data

Knowledge about how ecological networks vary across global scales is currently limited given the complexity of acquiring repeated spatial data for species interactions. Yet, recent developments in metawebs highlight efficient ways to first document possible interactions within regional species pools. Downscaling metawebs towards local network predictions is a promising approach to using the current data to investigate the variation of networks across space. However, issues remain in how to represent the spatial variability and uncertainty of species interactions, especially for large-scale food webs. Here, we present a probabilistic framework to downscale a metaweb based on the Canadian mammal metaweb and species occurrences from global databases. We investigated how our approach can be used to represent the variability of networks and communities between ecoregions in Canada. Species richness and interactions followed a similar latitudinal gradient across ecoregions but simultaneously identified contrasting diversity hotspots. Network motifs revealed additional areas of variation in network structure compared with species richness and number of links. Our method offers the potential to bring global predictions down to a more actionable local scale, and increases the diversity of ecological networks that can be projected in space. This article is part of the theme issue 'Connected interactions: enriching food web research by spatial and social interactions'.

Importance of plant and fruit traits on the structure of bird seed dispersal networks in different disturbed habitats

Species functional traits can influence seed dispersal processes and consequently affect species' role in the mutualistic network. Although the effect of animal traits on the structure of the seed dispersal network is well explored, it remains poorly understood how plant and fruit traits contribute to the structure. We here studied the effects of plant and fruit traits on the structure of bird seed dispersal networks across different disturbed habitats in the Meihua Mountain National Nature Reserve, Southeastern China. During the study period, 16, 20, 13, and 15 bird species were recorded foraging on 10, 11, 12, and 8 plant species, resulting in 511, 312, 265, and 201 foraging events in the protected forest, natural forest, village, and bamboo forest, respectively. The composition of these seed dispersal networks is not primarily influenced by a specific group of bulbul species, but rather by the presence of an endangered plant species, Taxus chinensis. As we expected, the structure of the four networks was different among the four disturbed habitats. Furthermore, our results also showed tree height and canopy density were the most important plant traits for structuring the seed dispersal network, while sugar, amylase, dry matter, and alkaloids were identified as significant fruit traits. Overall, our findings highlight the value of integrating trait-based ecology into the framework of the seed dispersal network and provide new insights for mutualistic network conservation in disturbed habitats.

Dissimilarity in flea and host assemblages and their interaction networks along a spatial distance gradient: different patterns revealed by different network dissimilarity metrics

We investigated the distance-decay pattern (an increase in dissimilarity with increasing geographic distance) in regional assemblages of fleas and their small mammalian hosts, as well as their interaction networks, in four biogeographic realms. Dissimilarity of assemblages (βtotal) was partitioned into species richness differences (βrich) and species replacement (βrepl) components. Dissimilarity of networks was assessed using two metrics: (a) whole network dissimilarity (βWN) partitioned into species replacement (βST) and interaction rewiring (βOS) components and (b) D statistics, measuring dissimilarity in the pure structure of the networks, without using information on species identities and calculated for hosts-shared-by-fleas networks (Dh) and fleas-shared-by-hosts networks (Df). We asked whether the distance-decay pattern (a) occurs among interactor assemblages or their interaction networks; (b) depends on the network dissimilarity metric used; and (c) differs between realms. The βtotal and βrepl of flea and host assemblages increased with distance in all realms except for host assemblages in the Afrotropics. βrich for flea and host assemblages increased with distance in the Nearctic only. In networks, βWN and βST demonstrated a distance-decay pattern, whereas βOS was mainly spatially invariant except in the Neotropics. Correlations of Dh or Df and geographic distance were mostly non-significant. We conclude that investigations of dissimilarity in interaction networks should include both types of dissimilarity metrics (those that consider partner identities and those that consider the pure structure of networks). This will allow elucidating the predictability of some facets of network dissimilarity and the unpredictability of other facets.

Pollen analysis reveals the effects of uncovered interactions, pollen-carrying structures, and pollinator sex on the structure of wild bee-plant networks

Pollination networks are increasingly used to model the complexity of interactions between pollinators and flowering plants in communities. Different methods exist to sample these interactions, with direct observations of plant-pollinator contacts in the field being by far the most common. Although the identification of pollen carried by pollinators allows uncovering interactions and increasing sample sizes, the methods used to build pollen-transport networks are variable and their effect on network structure remains unclear. To understand how interaction sampling influences the structure of networks, we analyzed the pollen found on wild bees from eight communities across Mallorca Island and investigated the differences in pollen loads between bee body parts (scopa vs. body) and sexes. We then assessed how these differences, as well as the uncovered interactions not detected in the field, influenced the structure of wild bee-plant networks. We identified a higher quantity and diversity of pollen in the scopa than in the rest of the female body, but these differences did not lead to differences in structure between plant-pollination (excluding scopa pollen) and bee-feeding interaction (including scopa pollen) networks. However, networks built with pollen data were richer in plant species and interactions and showed lower modularity and specialization (H2'), and higher nestedness than visitation networks based on field observations. Female interactions with plants were stronger compared to those of males, although not richer. Accordingly, females were more generalist (low d') and tended to be more central in interaction networks, indicating their more key role structuring pollination networks in comparison to males. Our study highlights the importance of palynological data to increase the resolution of networks, as well as to understand important ecological questions such as the differences between plant-pollination and bee-feeding interaction networks, and the role of sexes in pollination.

Predicting plant-pollinator interactions: concepts, methods, and challenges

Plant-pollinator interactions are ecologically and economically important, and, as a result, their prediction is a crucial theoretical and applied goal for ecologists. Although various analytical methods are available, we still have a limited ability to predict plant-pollinator interactions. The predictive ability of different plant-pollinator interaction models depends on the specific definitions used to conceptualize and quantify species attributes (e.g., morphological traits), sampling effects (e.g., detection probabilities), and data resolution and availability. Progress in the study of plant-pollinator interactions requires conceptual and methodological advances concerning the mechanisms and species attributes governing interactions as well as improved modeling approaches to predict interactions. Current methods to predict plant-pollinator interactions present ample opportunities for improvement and spark new horizons for basic and applied research.

Addition of nocturnal pollinators modifies the structure of pollination networks

Although the ecological network approach has substantially contributed to the study of plant-pollinator interactions, current understanding of their functional structure is biased towards diurnal pollinators. Nocturnal pollinators have been systematically ignored despite the publication of several studies that have tried to alleviate this diurnal bias. Here, we explored whether adding this neglected group of pollinators had a relevant effect on the overall architecture of three high mountain plant-pollinator networks. Including nocturnal moth pollinators modified network properties by decreasing total connectivity, connectance, nestedness and robustness to plant extinction; and increasing web asymmetry and modularity. Nocturnal moths were not preferentially connected to the most linked plants of the networks, and they were grouped into a specific "night" module in only one of the three networks. Our results indicate that ignoring the nocturnal component of plant-pollinator networks may cause changes in network properties different from those expected from random undersampling of diurnal pollinators. Consequently, the neglect of nocturnal interactions may provide a distorted view of the structure of plant-pollinator networks with relevant implications for conservation assessments.

Pollen DNA metabarcoding reveals cryptic diversity and high spatial turnover in alpine plant-pollinator networks

Alpine plant-pollinator communities play an important role in the functioning of alpine ecosystems, which are highly threatened by climate change. However, we still have a poor understanding of how environmental factors and spatiotemporal variability shape these communities. Here, we investigate what drives structure and beta diversity in a plant-pollinator metacommunity from the Australian alpine region using two approaches: pollen DNA metabarcoding (MB) and observations. Individual pollinators often carry pollen from multiple plant species, and therefore we expected MB to reveal a more diverse and complex network structure. We used two gene regions (ITS2 and trnL) to identify plant species present in the pollen loads of 154 insect pollinator specimens from three alpine habitats and construct MB networks, and compared them to networks based on observations alone. We compared species and interaction turnover across space for both types of networks, and evaluated their differences for plant phylogenetic diversity and beta diversity. We found significant structural differences between the two types of networks; notably, MB networks were much less specialized but more diverse than observation networks, with MB detecting many cryptic plant species. Both approaches revealed that alpine pollination networks are very generalized, but we estimated a high spatial turnover of plant species (0.79) and interaction rewiring (0.6) as well as high plant phylogenetic diversity (0.68) driven by habitat differences based on the larger diversity of plant species and species interactions detected with MB. Overall, our findings show that habitat and microclimatic heterogeneity drives diversity and fine-scale spatial turnover of alpine plant-pollinator networks.

What constrains food webs? A maximum entropy framework for predicting their structure with minimal biases

Food webs are complex ecological networks whose structure is both ecologically and statistically constrained, with many network properties being correlated with each other. Despite the recognition of these invariable relationships in food webs, the use of the principle of maximum entropy (MaxEnt) in network ecology is still rare. This is surprising considering that MaxEnt is a statistical tool precisely designed for understanding and predicting many types of constrained systems. This principle asserts that the least-biased probability distribution of a system's property, constrained by prior knowledge about that system, is the one with maximum information entropy. MaxEnt has been proven useful in many ecological modeling problems, but its application in food webs and other ecological networks is limited. Here we show how MaxEnt can be used to derive many food-web properties both analytically and heuristically. First, we show how the joint degree distribution (the joint probability distribution of the numbers of prey and predators for each species in the network) can be derived analytically using the number of species and the number of interactions in food webs. Second, we present a heuristic and flexible approach of finding a network's adjacency matrix (the network's representation in matrix format) based on simulated annealing and SVD entropy. We built two heuristic models using the connectance and the joint degree sequence as statistical constraints, respectively. We compared both models' predictions against corresponding null and neutral models commonly used in network ecology using open access data of terrestrial and aquatic food webs sampled globally (N = 257). We found that the heuristic model constrained by the joint degree sequence was a good predictor of many measures of food-web structure, especially the nestedness and motifs distribution. Specifically, our results suggest that the structure of terrestrial and aquatic food webs is mainly driven by their joint degree distribution.

metanetwork: A R package dedicated to handling and representing trophic metanetworks

Trophic networks describe interactions between species at a given location and time. Due to environmental changes, anthropogenic perturbations or sampling effects, trophic networks may vary in space and time. The collection of network time series or networks in different sites thus constitutes a metanetwork. We present here the R package metanetwork, which will ease the representation, the exploration and analysis of trophic metanetwork data sets that are increasingly available. Our main methodological advance consists in suitable layout algorithm for trophic networks, which is based on trophic levels and dimension reduction in a graph diffusion kernel. In particular, it highlights relevant features of trophic networks (trophic levels, energetic channels). In addition, we developed tools to handle, compare visually and quantitatively and aggregate those networks. Static and dynamic visualisation functions have been developed to represent large networks. We apply our package workflow to several trophic network data sets.

The reliability of regional ecological knowledge to build local interaction networks: a test using seed-dispersal networks across land-bridge islands

Building ecological networks is the fundamental basis of depicting how species in communities interact, but sampling complex interaction networks is extremely labour intensive. Recently, indirect ecological information has been applied to build interaction networks. Here we propose to extend the source of indirect ecological information, and applied regional ecological knowledge to build local interaction networks. Using a high-resolution dataset consisting of 22 locally observed networks with 17 572 seed-dispersal events, we test the reliability of indirectly derived local networks based on regional ecological knowledge (REK) across islands. We found that species richness strongly influenced 'local interaction rewiring' (i.e. the proportion of locally observed interactions among regionally interacting species), and all network properties were biased using REK-based networks. Notably, species richness and local interaction rewiring strongly affected estimations of REK-based network structures. However, locally observed and REK-based networks detected the same trends of how network structure correlates to island area and isolation. These results suggest that we should use REK-based networks cautiously for reflecting actual interaction patterns of local networks, but highlight that REK-based networks have great potential for comparative studies across environmental gradients. The use of indirect regional ecological information may thus advance our understanding of biogeographical patterns of species interactions.

Quantifying and estimating ecological network diversity based on incomplete sampling data

An ecological network refers to the ecological interactions among sets of species. Quantification of ecological network diversity and related sampling/estimation challenges have explicit analogues in species diversity research. A unified framework based on Hill numbers and their generalizations was developed to quantify taxonomic, phylogenetic and functional diversity. Drawing on this unified framework, we propose three dimensions of network diversity that incorporate the frequency (or strength) of interactions, species phylogenies and traits. As with surveys in species inventories, nearly all network studies are based on sampling data and thus also suffer from under-sampling effects. Adapting the sampling/estimation theory and the iNEXT (interpolation/extrapolation) standardization developed for species diversity research, we propose the iNEXT.link method to analyse network sampling data. The proposed method integrates the following four inference procedures: (i) assessment of sample completeness of networks; (ii) asymptotic analysis via estimating the true network diversity; (iii) non-asymptotic analysis based on standardizing sample completeness via rarefaction and extrapolation with network diversity; and (iv) estimation of the degree of unevenness or specialization in networks based on standardized diversity. Interaction data between European trees and saproxylic beetles are used to illustrate the proposed procedures. The software iNEXT.link has been developed to facilitate all computations and graphics. This article is part of the theme issue 'Detecting and attributing the causes of biodiversity change: needs, gaps and solutions'.

Shortcomings of reusing species interaction networks created by different sets of researchers

Given the requisite cost associated with observing species interactions, ecologists often reuse species interaction networks created by different sets of researchers to test their hypotheses regarding how ecological processes drive network topology. Yet, topological properties identified across these networks may not be sufficiently attributable to ecological processes alone as often assumed. Instead, much of the totality of topological differences between networks-topological heterogeneity-could be due to variations in research designs and approaches that different researchers use to create each species interaction network. To evaluate the degree to which this topological heterogeneity is present in available ecological networks, we first compared the amount of topological heterogeneity across 723 species interaction networks created by different sets of researchers with the amount quantified from non-ecological networks known to be constructed following more consistent approaches. Then, to further test whether the topological heterogeneity was due to differences in study designs, and not only to inherent variation within ecological networks, we compared the amount of topological heterogeneity between species interaction networks created by the same sets of researchers (i.e., networks from the same publication) with the amount quantified between networks that were each from a unique publication source. We found that species interaction networks are highly topologically heterogeneous: while species interaction networks from the same publication are much more topologically similar to each other than interaction networks that are from a unique publication, they still show at least twice as much heterogeneity as any category of non-ecological networks that we tested. Altogether, our findings suggest that extra care is necessary to effectively analyze species interaction networks created by different researchers, perhaps by controlling for the publication source of each network.

Analyzing trophic ecosystem functions with the interaction functional space

Quantifying the vulnerability of ecosystems to global change requires a better understanding of how trophic ecosystem functions emerge. So far, trophic ecosystem functions have been studied from the perspective of either functional diversity or network ecology. To integrate these two perspectives, we propose the interaction functional space (IFS) a conceptual framework to simultaneously analyze the effects of traits and interactions on trophic functions. We exemplify the added value of our framework for seed dispersal and wood decomposition and show how species interactions influence the relationship between functional trait diversity and trophic functions. We propose future applications for a range of functions where the IFS can help to elucidate mechanisms underpinning trophic functions and facilitate understanding of functional changes in ecosystems amidst global change.

Reciprocity and interaction effectiveness in generalised mutualisms among free-living species

Mutualistic interactions among free-living species generally involve low-frequency interactions and highly asymmetric dependence among partners, yet our understanding of factors behind their emergence is still limited. Using individual-based interactions of a super-generalist fleshy-fruited plant with its frugivore assemblage, we estimated the Resource Provisioning Effectiveness (RPE) and Seed Dispersal Effectiveness (SDE) to assess the balance in the exchange of resources. Plants were highly dependent on a few frugivore species, while frugivores interacted with most individual plants, resulting in strong asymmetries of mutual dependence. Interaction effectiveness was mainly driven by interaction frequency. Despite highly asymmetric dependences, the strong reliance on quantity of fruit consumed determined high reciprocity in rewards between partners (i.e. higher energy provided by the plant, more seedlings recruited), which was not obscured by minor variations in the quality of animal or plant service. We anticipate reciprocity will emerge in low-intimacy mutualisms where the mutualistic outcome largely relies upon interaction frequency.

Structure and characteristics of the plant-frugivore bird network from the Guilin Botanical Garden

The interaction between plants and frugivores is crucial to ecosystem function and community diversity. However, little is known about the interaction between plants and frugivorous bird species in urban green spaces. We observed interactions between plants and frugivorous birds in the Guilin Botanical Garden for one year and determined the structure and characteristics of the interaction network. We also analyzed the impact of species traits on their network roles. Interactions between 14 frugivorous birds and 13 fruit plant species were recorded in the study area. Autumn interactions comprised 38.79% of the overall network, and winter interactions comprised 33.15%. The modularity (Q, z-score) of the network was higher in autumn; the weighted nestedness (wNODF, z-score) and interaction evenness (E₂ , z-score) of the network were higher in winter; the connectance (C, z-score) and interaction diversity (z-score) of the network were higher in spring; and the specialization (H₂', z-score) of the network was higher in summer. The observed network showed lower C, lower interaction H₂ , lower E₂ , lower wNODF, higher H₂' and higher Q when compared to the random networks. The bird species most important to network stability were Hemixos castanonotus, Parus venustulus, and Pycnonotus sinensis. The most important plant species were Alocasia macrorrhiza, Cinnamomum camphora, and Machilus nanmu. Of all the bird and plant traits included in this study, only plant color had a significant impact on species strength, with black fruit having a higher species strength. Our results suggest that interaction networks in urban green spaces can be temporally complex and variable and that a network approach can be an important monitoring tool for detecting the status of crucial ecosystem functions.

Wing morphology is related to niche specialization and interaction networks in stenodermatine bats (Chiroptera: Phyllostomidae)

Plant–animal interactions constitute some of the most important ecological processes for the maintenance of tropical forests. Bats are the only group of mammals capable of true flight and have been recognized as important dispersers of pioneer and secondary successional plant species. Although progress has been made in the study of Neotropical bats, morphological variation of the wing and its influence on niche separation between species is unknown. We evaluated relationships among habitat structures of selected Colombian tropical dry forest patches, the diet through interaction networks, and wing morphology of 11 species of bats in the Stenodermatinae subfamily (297 individuals) using geometric morphometry in a phylogenetic context. The results indicate that the phylogenetic signal for wing size is greater than for wing shape, thus providing some evidence for evolutionary convergence. Wing shape variation was associated primarily with the distal anatomical tip of the third finger and the joint between the humerus and the radius and ulna. Species with wide, short wings, as in the genus Artibeus had generalist diets and less nested positions within the interaction networks. In contrast, species with elongated and pointed wings, such as Sturnira and Platyrrhinus, had specialized diets and more nested positions within the interaction networks. We argue that wing shape variation may play an important role as a source of interspecific variation leading to food specialization within tropical bat communities.

Global and regional ecological boundaries explain abrupt spatial discontinuities in avian frugivory interactions

Species interactions can propagate disturbances across space via direct and indirect effects, potentially connecting species at a global scale. However, ecological and biogeographic boundaries may mitigate this spread by demarcating the limits of ecological networks. We tested whether large-scale ecological boundaries (ecoregions and biomes) and human disturbance gradients increase dissimilarity among plant-frugivore networks, while accounting for background spatial and elevational gradients and differences in network sampling. We assessed network dissimilarity patterns over a broad spatial scale, using 196 quantitative avian frugivory networks (encompassing 1496 plant and 1004 bird species) distributed across 67 ecoregions, 11 biomes, and 6 continents. We show that dissimilarities in species and interaction composition, but not network structure, are greater across ecoregion and biome boundaries and along different levels of human disturbance. Our findings indicate that biogeographic boundaries delineate the world's biodiversity of interactions and likely contribute to mitigating the propagation of disturbances at large spatial scales.

Mechanisms underlying interaction frequencies and robustness in a novel seed dispersal network: lessons for restoration

As human-caused extinctions and invasions accumulate across the planet, understanding the processes governing ecological functions mediated by species interactions, and anticipating the effect of species loss on such functions become increasingly urgent. In seed dispersal networks, the mechanisms that influence interaction frequencies may also influence the capacity of a species to switch to alternative partners (rewiring), influencing network robustness. Studying seed dispersal interactions in novel ecosystems on O'ahu island, Hawai'i, we test whether the same mechanisms defining interaction frequencies can regulate rewiring and increase network robustness to simulated species extinctions. We found that spatial and temporal overlaps were the primary mechanisms underlying interaction frequencies, and the loss of the more connected species affected networks to a greater extent. Further, rewiring increased network robustness, and morphological matching and spatial and temporal overlaps between partners were more influential on network robustness than species abundances. We argue that to achieve self-sustaining ecosystems, restoration initiatives can consider optimal morphological matching and spatial and temporal overlaps between consumers and resources to maximize chances of native plant dispersal. Specifically, restoration initiatives may benefit from replacing invasive species with native species possessing characteristics that promote frequent interactions and increase the probability of rewiring (such as long fruiting periods, small seeds and broad distributions).

Data standardization of plant-pollinator interactions

Background

Animal pollination is an important ecosystem function and service, ensuring both the integrity of natural systems and human well-being. Although many knowledge shortfalls remain, some high-quality data sets on biological interactions are now available. The development and adoption of standards for biodiversity data and metadata has promoted great advances in biological data sharing and aggregation, supporting large-scale studies and science-based public policies. However, these standards are currently not suitable to fully support interaction data sharing.

Results

Here we present a vocabulary of terms and a data model for sharing plant–pollinator interactions data based on the Darwin Core standard. The vocabulary introduces 48 new terms targeting several aspects of plant–pollinator interactions and can be used to capture information from different approaches and scales. Additionally, we provide solutions for data serialization using RDF, XML, and DwC-Archives and recommendations of existing controlled vocabularies for some of the terms. Our contribution supports open access to standardized data on plant–pollinator interactions.

Conclusions

The adoption of the vocabulary would facilitate data sharing to support studies ranging from the spatial and temporal distribution of interactions to the taxonomic, phenological, functional, and phylogenetic aspects of plant–pollinator interactions. We expect to fill data and knowledge gaps, thus further enabling scientific research on the ecology and evolution of plant–pollinator communities, biodiversity conservation, ecosystem services, and the development of public policies. The proposed data model is flexible and can be adapted for sharing other types of interactions data by developing discipline-specific vocabularies of terms.

How network size strongly determines trophic specialisation: A technical comment on Luna et al. (2022)

Luna et al. (2022) concluded that the environment contributes to explaining specialisation in open plant-pollinator networks. When reproducing their study, we instead found that network size alone largely explained the variation in their specialisation metrics. Thus, we question whether empirical network specialisation is driven by the environment.

Worldwide Distribution of Antagonistic-Mutualistic Relationships Between Parrots and Palms

Palms, like all plants, show coevolutionary relationships with animals that have been traditionally categorized as mutualistic (seed dispersers and pollinators) or antagonistic (seed predators). This dual perspective, however, has prevented a full understanding of their true interactions with some animal groups, mainly those that do not ingest entire fruits. One clear example is parrots, which have been described to use palm species as feeding resources, while their role as seed dispersers has been largely neglected. Here, we combined fieldwork data with information from the literature and citizen science (i.e., naturalists and nature photographers) on parrot foraging ecology worldwide to evaluate the spatial and taxonomic extent of parrot-palm interactions and to identify the eco-evolutionary factors involved. We identified 1,189 interactions between 135 parrots and 107 palm species in more than 50 countries across the six realms where palms are present as natives or introduced. Combining this information, we identified 427 unique parrot-palm interacting pairs (i.e., a parrot species interacting with a palm species). Pure antagonistic interactions (i.e., parrots just preying on seeds or eating or destroying their non-reproductive parts) were less common (5%) than mutualistic ones (i.e., parrots benefiting by partially preying on the seed or fruit or consuming the pulp of the fruit or the flower but also contributing to seed dispersal and, potentially, pollination; 89%). After controlling for phylogeny, the size of consumed seeds and parrot body mass were positively related. Seed dispersal distances varied among palm species (range of estimated median dispersal distances: 9–250 m), with larger parrots dispersing seeds at greater distances, especially large fruits commonly categorized as megafauna anachronisms (>4 cm length). Although parrot-palm interactions are widespread, several factors (e.g., social behavior, predation fear, food availability, or seasonality) may affect the actual position of parrots on the antagonism-mutualism continuum for different palm species and regions, deserving further research. Meanwhile, the pervasiveness of parrot-palm mutualistic interactions, mainly involving seed dispersal and pollination, should not be overlooked in studies of palm ecology and evolution.

Response of avian and mammal seed dispersal networks to human-induced forest edges in a sub-humid tropical forest

Species-rich ecosystems as tropical forests are extremely vulnerable to anthropogenic destruction. Most tropical plant species rely on animals to disperse their seeds. However, patterns of mutualistic interactions have rarely been explored, and seed dispersal networks are still poorly studied in Africa. Here, we examine how forest edges’ (FE) seed dispersal networks differ from the mature forest (MF) at a West African sub-humid tropical forest within the National Park of Cantanhez (Guinea-Bissau). Additionally, we explore species’ roles within the network. MF had higher fruit availability, more frugivore visitors, and plant–frugivore interactions. Network structure was quite similar between habitats, showing signs of redundancy, and some robustness to species’ extinction. FE was more nested, modular, and specialized, whereas MF had higher connectance, interaction evenness, and robustness to extinction. Most species were generalists, but large-bodied frugivores prevailed at MF. FE showed a higher vulnerability, mostly to the loss of trees. Trees are key, keeping the structure of both networks. Large-bodied frugivores and fruiting-tree species that work as network connectors should thus be the focus of active conservation management in these forests. Only viable populations of these species will ensure a good performance of the seed dispersal network, promoting the natural regeneration of the ecosystem.

Unveiling the complexity and ecological function of aquatic macrophyte-animal networks in coastal ecosystems

Network theory offers innovative tools to explore the complex ecological mechanisms regulating species associations and interactions. Although interest in ecological networks has grown steadily during the last two decades, the application of network approaches has been unequally distributed across different study systems: while some kinds of interactions (e.g. plant-pollinator and host-parasite) have been extensively investigated, others remain relatively unexplored. Among the latter, aquatic macrophyte-animal associations in coastal environments have been largely neglected, despite their major role in littoral ecosystems. The ubiquity of macrophyte systems, their accessibility and multi-faceted ecological, economical and societal importance make macrophyte-animal systems an ideal subject for ecological network science. In fact, macrophyte-animal networks offer an aquatic counterpart to terrestrial plant-animal networks. In this review, we show how the application of network analysis to aquatic macrophyte-animal associations has the potential to broaden our understanding of how coastal ecosystems function. Network analysis can also provide a key to understanding how such ecosystems will respond to on-going and future threats from anthropogenic disturbance and environmental change. For this, we: (i) identify key issues that have limited the application of network theory and modelling to aquatic animal-macrophyte associations; (ii) illustrate through examples based on empirical data how network analysis can offer new insights on the complexity and functioning of coastal ecosystems; and (iii) provide suggestions for how to design future studies and establish this new research line into network ecology.

Addressing the Eltonian shortfall with trait-based interaction models

We have very limited knowledge of how species interact in most communities and ecosystems despite trophic relationships being fundamental for linking biodiversity to ecosystem functioning. A promising approach to fill this gap is to predict interactions based on functional traits, but many questions remain about how well we can predict interactions for different taxa, ecosystems and amounts of input data. Here, we built a new traits-based model of trophic interactions for European vertebrates and found that even models calibrated with 0.1% of the interactions (100 out of 71 k) estimated the full European vertebrate food web reasonably well. However, predators were easier to predict than prey, especially for some clades (e.g. fowl and storks) and local food web connectance was consistently overestimated. Our results demonstrate the ability to rapidly generate food webs when empirical data are lacking-an important step towards a more complete and spatially explicit description of food webs.

The effects of defaunation on plants' capacity to track climate change

Half of all plant species rely on animals to disperse their seeds. Seed dispersal interactions lost through defaunation and gained during novel community assembly influence whether plants can adapt to climate change through migration. We develop trait-based models to predict pairwise interactions and dispersal function for fleshy-fruited plants globally. Using interactions with introduced species as an observable proxy for interactions in future novel seed dispersal networks, we find strong potential to forecast their assembly and functioning. We conservatively estimate that mammal and bird defaunation has already reduced the capacity of plants to track climate change by 60% globally. This strong reduction in the ability of plants to adapt to climate change through range shifts shows a synergy between defaunation and climate change that undermines vegetation resilience.

Dietary abundance distributions: Dominance and diversity in vertebrate diets

Diet composition is among the most important yet least understood dimensions of animal ecology. Inspired by the study of species abundance distributions (SADs), we tested for generalities in the structure of vertebrate diets by characterising them as dietary abundance distributions (DADs). We compiled data on 1167 population-level diets, representing >500 species from six vertebrate classes, spanning all continents and oceans. DADs near-universally (92.5%) followed a hollow-curve shape, with scant support for other plausible rank-abundance-distribution shapes. This strong generality is inherently related to, yet incompletely explained by, the SADs of available food taxa. By quantifying dietary generalisation as the half-saturation point of the cumulative distribution of dietary abundance (sp50, minimum number of foods required to account for 50% of diet), we found that vertebrate populations are surprisingly specialised: in most populations, fewer than three foods accounted for at least half the diet. Variation in sp50 was strongly associated with consumer type, with carnivores being more specialised than herbivores or omnivores. Other methodological (sampling method and effort, taxonomic resolution), biological (body mass, frugivory) and biogeographic (latitude) factors influenced sp50 to varying degrees. Future challenges include identifying the mechanisms underpinning the hollow-curve DAD, its generality beyond vertebrates, and the biological determinants of dietary generalisation.

Combining DNA metabarcoding and ecological networks to inform conservation biocontrol by small vertebrate predators

In multifunctional landscapes, diverse communities of flying vertebrate predators provide vital services of insect pest control. In such landscapes, conservation biocontrol should benefit service-providing species to enhance the flow, stability and resilience of pest control services supporting the production of food and fiber. However, this would require identifying key service providers, which may be challenging when multiple predators interact with multiple pests. Here we provide a framework to identify the functional role of individual species to pest control in multifunctional landscapes. First, we used DNA metabarcoding to provide detailed data on pest species predation by diverse predator communities. Then, these data were fed into an extensive network analysis, in which information relevant for conservation biocontrol is gained from parameters describing network structure (e.g., modularity) and species roles in such network (e.g., centrality, specialization). We applied our framework to a Mediterranean landscape, where 19 bat species were found to feed on 132 insect pest species. Metabarcoding data revealed potentially important bats that consumed insect pest species in high frequency and/or diversity. Network analysis showed a modular structure, indicating sets of bat species that are required to regulate specific sets of insect pests. A few generalist bats had particularly important roles, either at network or module levels. Extinction simulations highlighted six bats, including species of conservation concern, which were sufficient to ensure that over three-quarters of the pest species had at least one bat predator. Combining DNA metabarcoding and ecological network analysis provides a valuable framework to identify individual species within diverse predator communities that might have a disproportionate contribution to pest control services in multifunctional landscapes. These species can be regarded as candidate targets for conservation biocontrol, although additional information is needed to evaluate their actual effectiveness in pest regulation.

Structural resilience and high interaction dissimilarity of plant-pollinator interaction networks in fire-prone grasslands

Fire is a frequent disturbance in most grasslands around the world, being key for the structure and dynamics of the biodiversity in such ecosystems. While grassland species may be resilient, little is known on how plant-pollinator networks reassemble after fire. Here, we investigate the structure and dynamics of plant-pollinator networks and the variation in species roles over a 2-year post-fire chronosequence on grassland communities in Southern Brazil. We found that both network specialization and modularity were similar over the chronosequence of time-since-fire, but in freshly burnt areas, there were more species acting as network hubs. Species roles exhibited high variation, with plant and pollinator species shifting roles along the post-disturbance chronosequence. Interaction dissimilarity was remarkably high in networks irrespective of times-since-fire. Interaction dissimilarity was associated more with rewiring than with species turnover, indicating that grassland plant and pollinator species are highly capable of switching partners. Time-since-fire had little influence on network structure but influenced the identity and diversity of pollinators playing key roles in the networks. These findings suggest that pollination networks in naturally fire-prone ecosystems are highly dynamic and resilient to fire with both plants and pollinators being highly capable of adjusting their interactions and network structure after disturbance.

A roadmap towards predicting species interaction networks (across space and time)

Networks of species interactions underpin numerous ecosystem processes, but comprehensively sampling these interactions is difficult. Interactions intrinsically vary across space and time, and given the number of species that compose ecological communities, it can be tough to distinguish between a true negative (where two species never interact) from a false negative (where two species have not been observed interacting even though they actually do). Assessing the likelihood of interactions between species is an imperative for several fields of ecology. This means that to predict interactions between species-and to describe the structure, variation, and change of the ecological networks they form-we need to rely on modelling tools. Here, we provide a proof-of-concept, where we show how a simple neural network model makes accurate predictions about species interactions given limited data. We then assess the challenges and opportunities associated with improving interaction predictions, and provide a conceptual roadmap forward towards predictive models of ecological networks that is explicitly spatial and temporal. We conclude with a brief primer on the relevant methods and tools needed to start building these models, which we hope will guide this research programme forward. This article is part of the theme issue 'Infectious disease macroecology: parasite diversity and dynamics across the globe'.

A plant-pollinator metanetwork along a habitat fragmentation gradient

To understand how plant-pollinator interactions respond to habitat fragmentation, we need novel approaches that can capture properties that emerge at broad scales, where multiple communities engage in metanetworks. Here we studied plant-pollinator interactions over 2 years on 29 calcareous grassland fragments selected along independent gradients of habitat size and surrounding landscape diversity of cover types. We associated network centrality of plant-pollinator interactions and grassland fragments with their ecological and landscape traits, respectively. Interactions involving habitat specialist plants and large-bodied pollinators were the most central, implying that species with these traits form the metanetwork core. Large fragments embedded in landscapes with high land cover diversity exhibited the highest centrality; however, small fragments harboured many unique interactions not found on larger fragments. Intensively managed landscapes have reached a point in which all remaining fragments matter, meaning that losing any further areas may vanish unique interactions with unknown consequences for ecosystem functioning.

Plant-pollinator conservation from the perspective of systems-ecology

Ecosystems are interconnected and complex, but conservation has often focused on rehabilitating individual species. A systems-ecology approach aims to support overall structure and maintain functions of the whole ecosystem, and may be especially pertinent for mutualistic plant-pollinator communities. This approach focuses on species interactions as the units to be conserved within the larger ecosystem. Analyzing species interactions is a more holistic approach because it incorporates a broader web of organisms, and considers the plethora of potential indirect influences from interacting partners. In this article, we suggest pollinator researchers focus on plant-pollinator networks to inform conservation programs and best support the coexistence of pollinators and plants within natural and agricultural systems. We propose that a system-ecology perspective is the most promising way to simultaneously improve pollinator conservation, agricultural sustainability, and human well-being.

Phenology and flowering overlap drive specialisation in plant-pollinator networks

Variation in dietary specialisation stems from fundamental interactions between species and their environment. Consequently, understanding the drivers of this variation is key to understanding ecological and evolutionary processes. Dietary specialisation in wild bees has received attention due to their close mutualistic dependence on plants, and because both groups are threatened by biodiversity loss. Many principles governing pollinator specialisation have been identified, but they remain largely unvalidated. Organismal phenology has the potential to structure realised specialisation by determining concurrent resource availability and pollinator foraging activity. We evaluate this principle using mechanistic models of adaptive foraging in pollinators within plant-pollinator networks. While temporal resource overlap has little impact on specialisation in pollinators with extended flight periods, reduced overlap increases specialisation as pollinator flight periods decrease. These results are corroborated empirically using pollen load data taken from bees with shorter and longer flight periods across environments with high and low temporal resource overlap.

Assessment of interactions between oil flowers and floral visitors in world biomes

Specialized glands that produce and store floral oil are found in 11 families of flowering plants, with greater representation in Malpighiaceae. Solitary bees use this resource to feed larvae and for cell lining. Oil interactions have been reported mainly at the population level and in some community studies that have commonly ignored not only the fewer representative families of oil-offering flowers, but also the floral visitors that do not collect floral oil. We have compiled and examined the interactions between oil-offering flowers and floral visitors in world biomes, to describe the structural patterns of the oil interaction meta-network and evaluate the species' functional roles. We reviewed 169 studies, which mainly used a phytocentric approach, reporting 1460 interactions between 214 species of oil-offering flowers and 377 species of floral visitors. Malpighiaceae and bees (mainly oil-collecting bees) were the most frequently investigated. The meta-network of oil interactions showed a modular structure, mainly influenced by botanical families, and core–periphery organization. The most important functional roles were associated with Malpighiaceae and oil-collecting bees, possibly due to their representativeness and geographic distribution.