Interaction strength in plant-pollinator networks: Are we using the right measure? (original) (raw)
Related papers
Ecology, 2019
Network analysis is a powerful tool to understand community-level plant-pollinator interactions. We evaluate the role of floral visitors on plant fitness through a series of pollination exclusion experiments to test the effectiveness of pollinators of an Ipomoea community in the Pacific coast of México, including: (1) all flower visitors, (2) visitors that contact the reproductive organs, (3) visitors that deposit pollen on stigmas, and (4) visitors that mediate fruit and seed production. Our results show that networks built from effective pollination interactions are smaller, less connected, more specialized and modular than floral visitor networks. Modules are associated with pollinator functional groups and they provide strong support for pollination syndromes only when non-effective interactions are excluded. In contrast to other studies, the analyzed networks are not nested. Our results also show that only 59% of floral visitors were legitimate pollinators that contribute to seed production. Furthermore, only 27% of the links in visitation network resulted in seed production. Our study shows that plant-pollination networks that consider effectiveness measures of pollination in addition to floral visitation provide insightful information about the different role floral visitors play in a community, encompassing a large number of commensalistic/antagonistic interactions and the more restricted set of mutualistic relationships that underlie the evolution of convergent floral phenotypes in plants.
Acta Oecologica
The decline of pollinators may alter the complex system of interactions that they establish with flowering plants, with potential negative consequences on both partners. Within this context, network analysis may be a useful tool to study ecological properties of plant-pollinator interactions and to evaluate the outcomes of conservation actions. Three conservation measures were implemented within the European LIFE+ PP-ICON project to support the local pollinator community of a population of the rare plant Dictamnus albus in a protected area near Bologna, Italy. Artificial nesting sites were installed to support solitary bees, populations of native plants were reinforced to increase foraging resources for pollinators, and colonies of bumblebees reared from wild queens were released in the study area. In this work we evaluate the effects of these conservation actions on plant-pollinator networks over a period of four years, comparing a pre-(2011-2012) and a post-conservation (2013-2014) action period. The overall network generalisation increased after the implementation of conservation measures and interactions were more evenly distributed. Module composition significantly changed between the two periods, showing a marked rewiring of interactions. D. albus was a module hub both before and after conservation actions, thus emerging as an important node within its own module. In addition, some plant and pollinator species directly targeted by conservation measures became module connectors, highlighting their increased importance in linking different modules. Finally, the reinforcement of plant and pollinator populations led to increased flower visitation. These results indicate that conservation actions affected species both directly and indirectly and that the network of interactions has potentially increased its robustness and resilience towards possible species loss. This study highlights ways in which network analysis can be used to measure changes in plant-pollinator interactions in response to conservation actions.
Ecology, 2018
Ecological interactions are highly dynamic in time and space. Previous studies of plant-animal mutualistic networks have shown that the occurrence of interactions varies substantially across years. We analyzed interannual variation of a quantitative mutualistic network, in which links are weighted by interaction frequency. The network was sampled over six consecutive years, representing one of the longest time series for a community-wide mutualistic network. We estimated the interannual similarity in interactions and assessed the determinants of their persistence. The occurrence of interactions varied greatly among years, with most interactions seen in only one year (64%) and few (20%) in more than two years. This variation was associated with the frequency and position of interactions relative to the network core, so that the network consisted of a persistent core of frequent interactions and many peripheral, infrequent interactions. Null model analyses suggest that species abundan...
Year‐to‐year variation in the topology of a plant–pollinator interaction network
2008
Vigorous discussion of the degree of specialization in pollination interactions, combined with advances in the analysis of complex networks, has revitalized the study of entire plantÁpollinator communities. Noticeably rare, however, are attempts to quantify temporal variation in the structure of plantÁpollinator networks, and to determine whether the status of species as specialists or generalists is stable.
The strength of plant-pollinator interactions
Ecology
Recent studies of plant-animal mutualistic networks have assumed that interaction frequency between mutualists predicts species impacts (population-level effects), and that field estimates of interaction strength (per-interaction effects) are unnecessary. Although existing evidence supports this assumption for the effect of animals on plants, no studies have evaluated it for the reciprocal effect of plants on animals. We evaluate this assumption using data on the reproductive effects of pollinators on plants and the reciprocal reproductive effects of plants on pollinators. The magnitude of species impacts of plants on pollinators, the reciprocal impacts of pollinators on plants, and their asymmetry were well predicted by interaction frequency. However, interaction strength was a key determinant of the sign of species impacts. These results underscore the importance of quantifying interaction strength in studies of mutualistic networks. We also show that the distributions of interaction strengths and species impacts are highly skewed, with few strong and many weak interactions. This skewed distribution matches the pattern observed in food webs, suggesting that the community-wide organization of species interactions is fundamentally similar between mutualistic and antagonistic interactions. Our results have profound ecological implications, given the key role of interaction strength for community stability.
Acta Oecologica, 2020
Most plant-pollinator networks are based on observations of contact between an insect and a flower in the field. Despite significant sampling efforts, some links are easier to report, while others remain unobserved. Therefore, visit-based networks represent a subsample of possible interactions in which the ignored part is variable. Pollen is a natural marker of insect visits to flowers. The identification of pollen found on insect bodies can be used as an alternative method to study plant-pollinator interactions, with a potentially lower risk of bias than the observation of visits, since it increases the number of interactions in the network. Here we compare plant-pollinator networks constructed (i) from direct observation of pollinator visits and (ii) from identification of pollen found on the same insects. We focused on three calcareous grasslands in France, with different plant and pollinator species diversities. Since pollen identification always yields richer, more connected networks, we focused our comparisons on sampling bias at equal network connectance. To do so, we first compared network structures with an analysis of latent blocks and motifs. We then compared species roles between both types of networks with an analysis of specialization and species positions within motifs. Our results suggest that the sampling from observations of insect visits does not lead to the construction of a network intrinsically different from the one obtained using pollen found on insect bodies, at least when field sampling strives to be exhaustive. Most of the significant differences are found at the species level, not at the network structure level, with singleton species accounting for a respectable fraction of these differences. Overall, this suggests that recording plant-pollinator interactions from pollinator visit observation does not provide a biased picture of the network structure, regardless of species richness; however, it provided less information on species roles than the pollen-based network.
Phenology determines the robustness of plant-pollinator networks
Scientific reports, 2018
Plant-pollinator systems are essential for ecosystem functioning, which calls for an understanding of the determinants of their robustness to environmental threats. Previous studies considering such robustness have focused mostly on species' connectivity properties, particularly their degree. We hypothesized that species' phenological attributes are at least as important as degree as determinants of network robustness. To test this, we combined dynamic modeling, computer simulation and analysis of data from 12 plant-pollinator networks with detailed information of topology of interactions as well as species' phenology of plant flowering and pollinator emergence. We found that phenological attributes are strong determinants of network robustness, a result consistent across the networks studied. Plant species persistence was most sensitive to increased larval mortality of pollinators that start earlier or finish later in the season. Pollinator persistence was especially se...
Ecology Letters, 2008
We analysed the dynamics of a plant-pollinator interaction network of a scrub community surveyed over four consecutive years. Species composition within the annual networks showed high temporal variation. Temporal dynamics were also evident in the topology of the network, as interactions among plants and pollinators did not remain constant through time. This change involved both the number and the identity of interacting partners. Strikingly, few species and interactions were consistently present in all four annual plant-pollinator networks (53% of the plant species, 21% of the pollinator species and 4.9% of the interactions). The high turnover in species-to-species interactions was mainly the effect of species turnover (c. 70% in pairwise comparisons among years), and less the effect of species flexibility to interact with new partners (c. 30%). We conclude that specialization in plant-pollinator interactions might be highly overestimated when measured over short periods of time. This is because many plant or pollinator species appear as specialists in 1 year, but tend to be generalists or to interact with different partner species when observed in other years. The high temporal plasticity in species composition and interaction identity coupled with the low variation in network structure properties (e.g. degree centralization, connectance, nestedness, average distance and network diameter) imply (i) that tight and specialized coevolution might not be as important as previously suggested and (ii) that plant-pollinator interaction networks might be less prone to detrimental effects of disturbance than previously thought. We suggest that this may be due to the opportunistic nature of plant and animal species regarding the available partner resources they depend upon at any particular time.
Interaction networks are widely used as tools to understand plant–pollinator communities, and to examine potential threats to plant diversity and food security if the ecosystem service provided by pollinating animals declines. However, most networks to date are based on recording visits to flowers, rather than recording clearly defined effective pollination events. Here we provide the first networks that explicitly incorporate measures of pollinator effectiveness (PE) from pollen deposition on stigmas per visit, and pollinator importance (PI) as the product of PE and visit frequency. These more informative networks, here produced for a low diversity heathland habitat, reveal that plant–pollinator interactions are more specialized than shown in most previous studies. At the studied site, the specialization index Embedded Image was lower for the visitation network than the PE network, which was in turn lower than Embedded Image for the PI network. Our study shows that collecting PE data is feasible for community-level studies in low diversity communities and that including information about PE can change the structure of interaction networks. This could have important consequences for our understanding of threats to pollination systems.
POLLINATION NETWORKS : GEOGRAPHIC PATTERNS conceptualization is in its very first phase (
Recent reviews of plant–pollinator mutualistic networks showed that generalization is a common pattern in this type of interaction. Here we examine the ecological correlates of generalization patterns in plant–pollinator networks, especially how interaction patterns covary with latitude, elevation, and insularity. We review the few published analyses of whole networks and include unpublished material, analyzing 29 complete plant– pollinator networks that encompass arctic, alpine, temperate, Mediterranean, and subtropical–tropical areas. The number of interactions observed (I) was a linear function of network size (M ) the maximum number of interactions: ln I 5 0.575 1 0.61 ln M; R2 5 0.946. The connectance (C), the fraction of observed interactions relative to the total possible, decreased exponentially with species richness, the sum of animal and plant species in each community (A 1 P): C 5 13.83 exp[20.003(A 1 P)]. After controlling for species richness, the residual connectance w...