Limited potential for bird migration to disperse plants to cooler latitudes.

Climate change is forcing the redistribution of life on Earth at an unprecedented velocity1,2. Migratory birds are thought to help plants to track climate change through long-distance seed dispersal3,4. However, seeds may be consistently dispersed towards cooler or warmer latitudes depending on whether the fruiting period of a plant species coincides with northward or southward migrations. Here we assess the potential of plant communities to keep pace with climate change through long-distance seed dispersal by migratory birds. To do so, we combine phenological and migration information with data on 949 seed-dispersal interactions between 46 bird and 81 plant species from 13 woodland communities across Europe. Most of the plant species (86%) in these communities are dispersed by birds migrating south, whereas only 35% are dispersed by birds migrating north; the latter subset is phylogenetically clustered in lineages that have fruiting periods that overlap with the spring migration. Moreover, the majority of this critical dispersal service northwards is provided by only a few Palaearctic migrant species. The potential of migratory birds to assist a small, non-random sample of plants to track climate change latitudinally is expected to strongly influence the formation of novel plant communities, and thus affect their ecosystem functions and community assembly at higher trophic levels.

Dietary metabarcoding reveals the simplification of bird-pest interaction networks across a gradient of agricultural cover.

Agriculture is vital for supporting human populations, but its intensification often leads to landscape homogenization and a decline in non-provisioning ecosystem services. Ecological intensification and multifunctional landscapes are suggested as nature-based alternatives to intensive agriculture, using ecological processes like natural pest regulation to maximize food production. Birds are recognized for their role in increasing crop yields by consuming invertebrate pests in several agroecosystems. However, the understanding of how bird species, their traits and agricultural land cover influence the structure of bird-pest interactions remains limited. We sampled bird-pest interactions monthly for 1 year, at four sites within a multifunctional landscape, following a gradient of increasing agricultural land cover. We analysed 2583 droppings of 55 bird species with DNA metabarcoding and detected 225 pest species in 1139 samples of 42 bird species. As expected, bird-pest interactions were highly variable across bird species. Dietary pest richness was lower in the fully agricultural site, while predation frequency remained consistent across the agricultural land cover gradient. Network analysis revealed a reduction in the complexity of bird-pest interactions as agricultural coverage increased. Bird species abundance affected the bird's contribution to the network structure more than any of the bird traits analysed (weight, phenology, invertebrate frequency in diet and foraging strata), with more common birds being more important to network structure. Overall, our results show that increasing agricultural land cover increases the homogenization of bird-pest interactions. This shows the importance of maintaining natural patches within agricultural landscapes for biodiversity conservation and enhanced biocontrol.

A agricultura é essencial para suportar a população humana, mas a sua intensificação geralmente leva à homogeneização da paisagem e à redução dos serviços do ecossistema que não sejam de provisão. A intensificação ecológica e paisagens multifuncionais são sugeridas como alternativas naturais à agricultura intensiva, utilizando processos ecológicos como a regulação natural de pragas para maximizar a produção de alimentos. As aves são conhecidas pelo seu papel no aumento da produtividade das culturas por consumirem pragas em diversos agroecossistemas. Contudo, o conhecimento de como as espécies de aves, as suas características e a cobertura agrícola influenciam as interações entre aves e pragas são limitados. Nós amostrámos estas interações mensalmente durante um ano, em quatro locais, numa paisagem multifuncional, ao longo um gradiente de aumento da cobertura agrícola. Analisamos 2583 dejetos de 55 espécies de aves com DNA metabarcoding e detetamos 225 espécies praga em 1139 amostras de 42 espécies de aves. Como esperado, as interações entre aves e pragas foram muito distintas entre as várias espécies de aves. A riqueza de pragas na dieta foi menor no local completamente dominado por área agrícola, enquanto a frequência de predação de pragas foi constante ao longo do gradiente de cobertura agrícola. A análise de redes demonstrou uma redução na complexidade das interações entre aves e pragas à medida que a cobertura agrícola aumenta. A abundância das espécies de aves influenciou mais a contribuição das aves para a estrutura da rede do que qualquer uma das características analisadas (peso, fenologia, frequência de invertebrados na dieta e estrato de alimentação), sendo as aves mais comuns as mais importantes na estrutura da rede. De forma geral, os nossos resultados indicam que o aumento da cobertura agrícola aumenta a homogeneização das interações entre aves e pragas. Isto demonstra a importância de preservar áreas naturais em paisagem agrícolas para a conservação de biodiversidade e melhor controlo biológico.

Pollination networks from natural and anthropogenic-novel communities show high structural similarity.

The Anthropocene is marked by an unprecedented homogenisation of the world's biota, confronting species that never co-occurred during their evolutionary histories. Interactions established in these novel communities may affect ecosystem functioning; however, most research has focused on the impacts of a minority of aggressive invasive species, while changes inflicted by a less conspicuous majority of non-invasive alien species on community structure are still poorly understood. This information is critical to guide conservation strategies, and instrumental to advance ecological theory, particularly to understand how non-native species integrate in recipient communities and affect the interactions of native species. We evaluated how the structure of 50 published pollination networks changes with the proportion of alien plant species and found that network structure is largely unaffected. Although some communities were heavily invaded, the proportion of alien plant species was relatively low (mean = 10%; max. = 38%). We further characterized the pollination network in a botanic garden with a plant community dominated by non-invasive alien species (85%). We show that the structure of this novel community is also not markedly different from native-dominated communities. Plant-pollinator interactions revealed no obvious differences regarding plant origin (native vs. alien) or the native bioregion of the introduced plants. This overall similarity between native and alien plants is likely driven by the contrasting patterns of invasive plants (promoting generalism), and non-invasive aliens, suggested here to promote specialization.

Four-trophic level food webs reveal the cascading impacts of an invasive plant targeted for biocontrol.

Biological invasions are a major threat to biodiversity and as such understanding their impacts is a research priority. Ecological networks provide a valuable tool to explore such impacts at the community level, and can be particularly insightful for planning and monitoring biocontrol programmes, including the potential for their seldom evaluated indirect non-target effects. Acacia longifolia is among the worst invasive species in Portugal, and has been recently targeted for biocontrol by a highly specific gall-wasp. Here we use an ambitious replicated network approach to: (1) identify the mechanisms by which direct and indirect impacts of A. longifolia can cascade from plants to higher trophic levels, including gallers, their parasitoids and inquilines; (2) reveal the structure of the interaction networks between plants, gallers, parasitoids and inquilines before the biocontrol; and (3) explore the potential for indirect interactions among gallers, including those established with the biocontrol agent, via apparent competition. Over a 15-month period, we collected 31,737 galls from native plants and identified all emerging insects, quantifying the interactions between 219 plant-, 49 galler-, 65 parasitoid- and 87 inquiline-species-one of the largest ecological networks to date. No galls were found on any of the 16 alien plant species. Invasion by A. longifolia caused an alarming simplification of plant communities, with cascading effects to higher trophic levels, namely: a decline of overall gall biomass, and on the richness, abundance and biomass of galler insects, their parasitoids, and inquilines. Correspondingly, we detected a significant decline in the richness of interactions between plants and galls. The invasion tended to increase overall interaction evenness by promoting the local extinction of the native plants that sustained more gall species. However, highly idiosyncratic responses hindered the detection of further consistent changes in network topology. Predictions of indirect effects of the biocontrol on native gallers via apparent competition ranged from negligible to highly significant. Such scenarios are incredibly hard to predict, but even if there are risks of indirect effects it is critical to weigh them carefully against the consequences of inaction and invasive species spread.

Anthropogenic forests simplify seed- but not pollen-dispersal networks.

Natural native forests are rapidly being replaced by anthropogenic forests often with a strong presence of invasive alien plant species. Eucalypt species are widely planted worldwide, with Eucalyptus globulus plantations being particularly expressive in Portugal. Poor forestry practices often lead to the associated expansion of invasive species, such as Acacia dealbata. However, we still know relatively little about the functioning of anthropogenic forests, such as seed and pollen dispersal services. Here, we compared bird abundance and richness and the seed and pollen dispersal networks in both forest types. Anthropogenic forests presented lower bird abundance, and smaller, more simplified, and more random (abundance-based) seed dispersal services than those of natural forests. Interestingly, the pollen dispersal network was more similar than the seed dispersal network for both forest types and dominated by opportunistic and neutral processes, given the absence of specialized nectarivorous. The proportion of birds transporting seeds decreased, while those carrying pollen significantly increased in the anthropogenic forest compared to the native forest. Our work highlights the impact of anthropogenic forests on bird abundance, with consequences for seed dispersal services and forest regeneration.

Seed dispersal networks in the Galápagos and the consequences of alien plant invasions.

Alien plants are a growing threat to the Galápagos unique biota. We evaluated the impact of alien plants on eight seed dispersal networks from two islands of the archipelago. Nearly 10 000 intact seeds from 58 species were recovered from the droppings of 18 bird and reptile dispersers. The most dispersed invaders were Lantana camara, Rubus niveus and Psidium guajava, the latter two likely benefiting from an asynchronous fruit production with most native plants, which facilitate their consumption and spread. Lava lizards dispersed the seeds of 27 species, being the most important dispersers, followed by small ground finch, two mockingbirds, the giant tortoise and two insectivorous birds. Most animals dispersed alien seeds, but these formed a relatively small proportion of the interactions. Nevertheless, the integration of aliens was higher in the island that has been invaded for longest, suggesting a time-lag between alien plant introductions and their impacts on seed dispersal networks. Alien plants become more specialized with advancing invasion, favouring more simplified plant and disperser communities. However, only habitat type significantly affected the overall network structure. Alien plants were dispersed via two pathways: dry-fruited plants were preferentially dispersed by finches, while fleshy fruited species were mostly dispersed by other birds and reptiles.

Effects of alien plants on insect abundance and biomass: a food-web approach.

The replacement of native plants by alien species is likely to affect other trophic levels, particularly phytophagous insects. Nevertheless, the effect of alien plants on insect biomass has not yet been quantified. Given their critical role in transferring energy from plants to higher trophic levels, if alien plants do affect insect biomass, this could have far-reaching consequences for community structure. We used 35 food webs to evaluate the impacts of alien plants on insect productivity in a native forest in the Azores. Our food webs quantified plants, insect herbivores, and their parasitoids, which allowed us to test the effects of alien plants on species richness and evenness, insect abundance, insect biomass, and food-web structure. Species richness of plants and insects, along with plant species evenness, declined as the level of plant invasion increased. Nevertheless, none of the 4 quantitative food-web descriptors (number of links, link density, connectance, and interaction evenness) varied significantly with plant invasion independent of the size of the food web. Overall, insect abundance was not significantly affected by alien plants, but insect biomass was significantly reduced. This effect was due to the replacement of large insects on native plants with small insects on alien plants. Furthermore, the impact of alien plants was sufficiently severe to invert the otherwise expected pattern of species-richness decline with increased elevation. We predict a decrease in insect productivity by over 67% if conservation efforts fail to halt the invasion of alien plants in the Azores.

Disclosing the double mutualist role of birds on Galápagos.

Life on oceanic islands deviate in many ways from that on the mainland. Their biodiversity is relatively poor and some groups are well-represented, others not, especially not insects. A scarcity of insects forces birds to explore alternative food, such as nectar and fruit. In this way, island birds may pollinate and disperse seed to an extent unseen on any mainland; they may even first consume floral resources of a plant species and then later harvest the fruit of the same species. Through this biotic reuse, they may act as double mutualists. The latter have never been studied at the level of the network, because they are traditionally considered rare. We sampled pollination and seed-dispersal interactions on Galápagos and constructed a plant-bird mutualism network of 108 plant (12% being double mutualists) and 21 bird species (48% being double mutualists), and their 479 interactions, being either single (95%) or double mutualisms (5%). Double mutualists constitute the core in the pollination-dispersal network, coupling the two link types together. They may also initiate positive feedbacks (more pollination leading to more dispersal), which theoretically are known to be unstable. Thus, double mutualisms may be a necessary, but risky prerequisite to the survival of island biodiversity.

Long-distance dispersal to oceanic islands: success of plants with multiple diaspore specializations.

A great number of scientific papers claim that angiosperm diversification is manifested by an ample differentiation of diaspore traits favouring long-distance seed dispersal. Oceanic islands offer an ideal framework to test whether the acquisition of multiple sets of diaspore traits (syndromes) by a single species results in a wider geographic distribution. To this end, we performed floristic and syndrome analyses and found that diplochorous species (two syndromes) are overrepresented in the recipient flora of the Azores in contrast to that of mainland Europe, but not to mainland Portugal. An additional analysis of inter-island colonization showed a general trend of a higher number of islands colonized by species with a single syndrome (monochorous) and two syndromes than species with no syndrome (unspecialized). Nevertheless, statistical significance for differences in colonization is meagre in some cases, partially due to the low proportion of diplochorous species in Europe (244 of ∼10 000 species), mainland Portugal (89 of 2294 species), and the Azores (9 of 148 species), Canaries (17 of 387 lowland species) and Galápagos (18 of 313 lowland species). Contrary to expectations, this first study shows only a very marginal advantage for long-distance dispersal of species bearing multiple syndromes.

Historical isolation of the Galápagos carpenter bee (Xylocopa darwini) despite strong flight capability and ecological amplitude.

Colonization across the Galápagos Islands by the carpenter bee (Xylocopa darwini) was reconstructed based on distribution of mitochondrial haplotypes (cytochrome oxidase II (COII) sequences) and haplotype lineages. A total of 12 haplotypes were found in 118 individuals of X. darwini. Distributional, phylogenetic and phylogeographic analyses suggest early colonization of most islands followed by historical isolation in two main groups: eastern and central-western islands. Evidence of recurrent inter-island colonization of haplotypes is largely lacking, despite strong flight capability and ecological amplitude of the species. Recent palaeogeographic data suggest that several of the current islands were connected in the past and thus the isolation pattern may have been even more pronounced. A contrast analysis was also carried out on 10 animal groups of the Galápagos Islands, and on haplotype colonization of seven animal and plant species from several oceanic archipelagos (the Galápagos, Azores, Canary Islands). New colonization metrics on the number of potential vs. inferred colonization events revealed that the Galápagos carpenter bee shows one of the most significant examples of geographic isolation.