Honey bees (Apis mellifera) modify plant-pollinator network structure, but do not alter wild species' interactions.

Honey bees (Apis mellifera) are widely used for honey production and crop pollination, raising concern for wild pollinators, as honey bees may compete with wild pollinators for floral resources. The first sign of competition, before changes appear in wild pollinator abundance or diversity, may be changes to wild pollinator interactions with plants. Such changes for a community can be measured by looking at changes to metrics of resource use overlap in plant-pollinator interaction networks. Studies of honey bee effects on plant-pollinator networks have usually not distinguished whether honey bees alter wild pollinator interactions, or if they merely alter total network structure by adding their own interactions. To test this question, we experimentally introduced honey bees to a Canadian grassland and measured plant-pollinator interactions at varying distances from the introduced hives. We found that honey bees increased the network metrics of pollinator and plant functional complementarity and decreased interaction evenness. However, in networks constructed from just wild pollinator interactions, honey bee abundance did not affect any of the metrics calculated. Thus, all network structural changes to the full network (including honey bee interactions) were due only to honey bee-plant interactions, and not to honey bees causing changes in wild pollinator-plant interactions. Given widespread and increasing use of honey bees, it is important to establish whether they affect wild pollinator communities. Our results suggest that honey bees did not alter wild pollinator foraging patterns in this system, even in a year that was drier than the 20-year average.

Correction: Predicting direct and indirect non-target impacts of biocontrol agents using machine-learning approaches.

[This corrects the article DOI: 10.1371/journal.pone.0252448.].

Predicting direct and indirect non-target impacts of biocontrol agents using machine-learning approaches.

Biological pest control (i.e. 'biocontrol') agents can have direct and indirect non-target impacts, and predicting these effects (especially indirect impacts) remains a central challenge in biocontrol risk assessment. The analysis of ecological networks offers a promising approach to understanding the community-wide impacts of biocontrol agents (via direct and indirect interactions). Independently, species traits and phylogenies have been shown to successfully predict species interactions and network structure (alleviating the need to collect quantitative interaction data), but whether these approaches can be combined to predict indirect impacts of natural enemies remains untested. Whether predictions of interactions (i.e. direct effects) can be made equally well for generalists vs. specialists, abundant vs. less abundant species, and across different habitat types is also untested for consumer-prey interactions. Here, we used two machine-learning techniques (random forest and k-nearest neighbour; KNN) to test whether we could accurately predict empirically-observed quantitative host-parasitoid networks using trait and phylogenetic information. Then, we tested whether the accuracy of machine-learning-predicted interactions depended on the generality or abundance of the interacting partners, or on the source (habitat type) of the training data. Finally, we used these predicted networks to generate predictions of indirect effects via shared natural enemies (i.e. apparent competition), and tested these predictions against empirically observed indirect effects between hosts. We found that random-forest models predicted host-parasitoid pairwise interactions (which could be used to predict attack of non-target host species) more successfully than KNN. This predictive ability depended on the generality of the interacting partners for KNN models, and depended on species' abundances for both random-forest and KNN models, but did not depend on the source (habitat type) of data used to train the models. Further, although our machine-learning informed methods could significantly predict indirect effects, the explanatory power of our machine-learning models for indirect interactions was reasonably low. Combining machine-learning and network approaches provides a starting point for reducing risk in biocontrol introductions, and could be applied more generally to predicting species interactions such as impacts of invasive species.

Using Network Theory to Understand and Predict Biological Invasions.

Understanding and predicting biological invasions is challenging because of the complexity of many interacting players. A holistic approach is needed with the potential to simultaneously consider all relevant effects and effectors. Using networks to describe the relevant anthropogenic and ecological factors, from community-level to global scales, promises advances in understanding aspects of invasion from propagule pressure, through establishment, spread, and ecological impact of invaders. These insights could lead to development of new tools for prevention and management of invasions that are based on species' network characteristics and use of networks to predict the ecological effects of invaders. Here, we review the findings from network ecology that show the most promise for invasion biology and identify pressing needs for future research.

Non-random food-web assembly at habitat edges increases connectivity and functional redundancy.

Habitat fragmentation dramatically alters the spatial configuration of landscapes, with the creation of artificial edges affecting community structure and dynamics. Despite this, it is not known how the different food webs in adjacent habitats assemble at their boundaries. Here we demonstrate that the composition and structure of herbivore-parasitoid food webs across edges between native and plantation forests are not randomly assembled from those of the adjacent communities. Rather, elevated proportions of abundant, interaction-generalist parasitoid species at habitat edges allowed considerable interaction rewiring, which led to higher linkage density and less modular networks, with higher parasitoid functional redundancy. This was despite high overlap in host composition between edges and interiors. We also provide testable hypotheses for how food webs may assemble between habitats with lower species overlap. In an increasingly fragmented world, non-random assembly of food webs at edges may increasingly affect community dynamics at the landscape level.

Apparent competition drives community-wide parasitism rates and changes in host abundance across ecosystem boundaries.

Species have strong indirect effects on others, and predicting these effects is a central challenge in ecology. Prey species sharing an enemy (predator or parasitoid) can be linked by apparent competition, but it is unknown whether this process is strong enough to be a community-wide structuring mechanism that could be used to predict future states of diverse food webs. Whether species abundances are spatially coupled by enemy movement across different habitats is also untested. Here, using a field experiment, we show that predicted apparent competitive effects between species, mediated via shared parasitoids, can significantly explain future parasitism rates and herbivore abundances. These predictions are successful even across edges between natural and managed forests, following experimental reduction of herbivore densities by aerial spraying of insecticide over 20 hectares. This result shows that trophic indirect effects propagate across networks and habitats in important, predictable ways, with implications for landscape planning, invasion biology and biological control.

Persist or Produce: A Community Trade-Off Tuned by Species Evenness.

Understanding the effects of biodiversity on community persistence and productivity is key to managing both natural and production systems. Because rare species face greater danger of extinction, species evenness, a measure of how similar abundances are across species in a community, is seen as a key component of biodiversity. However, previous studies have failed to find a consistent association of species evenness with species survival and biomass production. Here we provide a theoretical framework for the relationship among these three elements. We demonstrate that the lack of consistent outcomes is not an idiosyncratic artifact of different studies but can be unified under one common framework. Applying a niche theory approach, we confirm that under demographic stochasticity evenness is a general indicator of the risk of future species extinctions in a community, in accordance with the majority of empirical studies. In contrast, evenness cannot be used as a direct indicator of the level of biomass production in a community. When a single species dominates, as expressed by the constraints imposed by the population dynamics, biomass production depends on the niche position of the dominating species and can increase or decrease with evenness. We demonstrate that high species evenness and an intermediate level of biomass production is the configuration that maximizes the average species survival probability in response to demographic stochasticity.

Community-level net spillover of natural enemies from managed to natural forest.

Edge effects in fragmented natural habitats may De exaceroateci by intensive land use in the surrounding landscape. Given that most managed systems have higher primary productivity than adjacent natural systems, theory suggests that bottom-up subsidized consumers are likely to spill over from managed to natural habitats. Furthermore, the magnitude of spillover is likely to differ between generalist and specialist consumers, because of differences in their ability to use the full spectrum of resources. However, it is unknown whether there is indeed asymmetrical spillover of consumers between managed and natural habitats, and whether this is related to resource abundance or the trophic specialization of the consumer. We used flight intercept traps to measure spillover of generalist predators (Vespula wasps, Vespidae) and more specialist predators (106 species of parasitoids, Ichneumonidae and Braconidae) across habitat edges between native New Zealand forest and exotic plantation forest over a summer season. We found net spillover of both generalist and specialist predators from plantation to native forest, and that this was greater for generalists. To test whether natural enemy spillover from managed habitats was related to prey (caterpillar) abundance (i.e., whether it was bottom-up productivity driven, due to increased primary productivity), we conducted a large-scale herbivore reduction experiment at half of our plantation sites, by helicopter spraying caterpillar-specific insecticide over 2.5 ha per site. We monitored bidirectional natural enemy spillover and found that herbivore reduction reduced generalist but not specialist predator spillover. Trophic generalists may benefit disproportionately from high resource productivity in a habitat, and their cross-habitat spillover effects on natural food webs may be an important source of consumer pressure in mosaic landscapes.

Phylogenetic diversity and co-evolutionary signals among trophic levels change across a habitat edge.

Incorporating the evolutionary history of species into community ecology enhances understanding of community composition, ecosystem functioning and responses to environmental changes. Phylogenetic history might partly explain the impact of fragmentation and land-use change on assemblages of interacting organisms and even determine potential cascading effects across trophic levels. However, it remains unclear whether phylogenetic diversity of basal resources is reflected at higher trophic levels in the food web. In particular, phylogenetic determinants of community structure have never been incorporated into habitat edge studies, even though edges are recognized as key factors affecting communities in fragmented landscapes. Here, we test whether phylogenetic diversity at different trophic levels (plants, herbivores and parasitoids) and signals of co-evolution (i.e. phylogenetic congruence) among interacting trophic levels change across an edge gradient between native and plantation forests. To ascertain whether there is a signal of co-evolution across trophic levels, we test whether related consumer species generally feed on related resource species. We found differences across trophic levels in how their phylogenetic diversity responded to the habitat edge gradient. Plant and native parasitoid phylogenetic diversity changed markedly across habitats, while phylogenetic variability of herbivores (which were predominantly native) did not change across habitats, though phylogenetic evenness declined in plantation interiors. Related herbivore species did not appear to feed disproportionately on related plant species (i.e. there was no signal of co-evolution) even when considering only native species, potentially due to the high trophic generality of herbivores. However, related native parasitoid species tended to feed on related herbivore species, suggesting the presence of a co-evolutionary signal at higher trophic levels. Moreover, this signal was stronger in plantation forests, indicating that this habitat may impose stresses on parasitoids that constrain them to attack only host species for which they are best adapted. Overall, changes in land use across native to plantation forest edges differentially affected phylogenetic diversity across trophic levels, and may also exert a strong selective pressure for particular co-evolved herbivore-parasitoid interactions.

Complementarity and redundancy of interactions enhance attack rates and spatial stability in host-parasitoid food webs.

Complementary resource use and redundancy of species that fulfill the same ecological role are two mechanisms that can respectively increase and stabilize process rates in ecosystems. For example, predator complementarity and redundancy can determine prey consumption rates and their stability, yet few studies take into account the multiple predator species attacking multiple prey at different rates in natural communities. Thus, it remains unclear whether these biodiversity mechanisms are important determinants of consumption in entire predator-prey assemblages, such that food-web interaction structure determines community-wide consumption and stability. Here, we use empirical quantitative food webs to study the community-wide effects of functional complementarity and redundancy of consumers (parasitoids) on herbivore control in temperate forests. We find that complementarity in host resource use by parasitoids was a strong predictor of absolute parasitism rates at the community level and that redundancy in host-use patterns stabilized community-wide parasitism rates in space, but not through time. These effects can potentially explain previous contradictory results from predator diversity research. Phylogenetic diversity (measured using taxonomic distance) did not explain functional complementarity or parasitism rates, so could not serve as a surrogate measure for functional complementarity. Our study shows that known mechanisms underpinning predator diversity effects on both functioning and stability can easily be extended to link food webs to ecosystem functioning.