P.01 Accumulation and effects of the neonicotinoid insecticide clothianidin in bumblebees (Bombus terrestris)
Malin Røyset Aarønes University of Oslo, Norway
Neonicotinoids can cause lethal and sublethal effects in bumblebees, with exposure trough pollen and nectar collected from treated crops. However, the bioaccumulative potential of these neonicotinoids has not been studied before. Our study aimed to assess the accumulation of clothianidin, a neonicotinoid, in bumblebee workers and the queen, and assess whether the accumulation could cause a change in mortality, brood production, nectar consumption, and storage of food. Bumblebee colonies (Bombus terrestris, n = 48) were exposed to field-realistic concentrations of clothianidin through nectar, with concentrations ranging from 1 μg/L to 13 μg/L, in a chronic exposure regime lasting nine days. Clothianidin showed a dose-response accumulation in the head (<0.2 – 2.17 μg/kg) and body (<0.2 – 3.17 μg/kg) of workers, and in the body (<0.2 – 2.49 μg/kg) of the queen, although the concentration was below that measured in the nectar (BAF = 0.2). Exposure did not cause a change in mortality or brood production, but showed a trend of a hormeotic response in nectar consumption and a negative dose-response trend in the proportion of empty honeypots. Accumulation in the head and body, as well as changes in nectar consumption and food storage, have the potential to cause long-term detrimental effects on brood production and mortality.
P.02 Pollinators enhance crop yield and shorten the growing season by modulating plant functional traits: A comparison of 23 canola varieties
George C. Adamidis1,2, Ralph V. Cartar1, Andony P. Melathopoulos3, Stephen F. Pernal4, Shelley E. Hoover5 1 Department of Biological Sciences, University of Calgary, Calgary, Canada 2 Institute of Ecology and Evolution, University of Bern, Switzerland 3 Department of Horticulture, Oregon State University, Corvallis, OR, United States of America 4 Agriculture and Agri-Food Canada, Beaverlodge Research Farm, Beaverlodge, AB, Canada 5 Alberta Agriculture and Forestry, Lethbridge, AB, Canada
Insect pollination of flowers should change the within-season allocation of resources in plants. But the nature of this life-history response, particularly regarding allocation to roots, photosynthetic structures, and flowers, is empirically unresolved. This study uses a greenhouse experiment to investigate the effect of insect pollination on the yield of 23 varieties of a globally important crop—canola (Brassica napus). Overall, insect pollination modified the functional characteristics (flower timing, flower effort, plant size & shape, seed packaging, root biomass) of canola crops, increasing yield quantity and quality, and pollinator dependence. Yield and pollinator dependence were defined by strong trait trade-offs, which ranged from more pollinator-dependent plants favouring early reproductive effort, to less pollinator-dependent canola plants favouring a prolonged phenology with smaller plant size and lower seed quality. Yield decreased with pollinator dependence in the absence of pollinators. The current preference for hybrid varieties will increase yield compared to open-pollinated varieties, but, even so, pollinators typically enhance yield of both. Our study elucidates the mechanisms through which insect pollination alters the character and function of a globally important crop, supporting optimization of yield via intensification of insect pollination, and highlights the beneficial (fitness) effects of insect pollination early in the season.
P.03 Herbivory-induced reduction on community-wide pollination services
Luis Aguirre University of Massachusetts, Amherst, United States of America
Pollination is an ecological phenomenon of great importance in natural communities, but little is known about the effects that trophic interactions have on this process at the community level. In many plants, deposition of conspecific pollen is imperative for reproduction and is tied directly to the behavior of their pollinators. In turn, pollinator behavior is determined by floral traits. However, many floral traits often respond negatively to herbivory attacks. The aims of this project are to determine how herbivory to a dominant plant species affects pollinator behavior, and how these changes affect reproduction in undamaged neighboring species. Here, we performed a natural experiment in a milkweed-dominated community in which we simulated herbivory to milkweeds by removing foliar tissue and applying jasmonic acid. We compared pollinator behavior in herbivory and control plots by analyzing pollen loads of floral visitors as to determine degrees of floral fidelity. We also measured the quality of pollination services to undamaged neighboring plants by assessing pollen deposition and seed production in undamaged neighboring plant species. Our results indicate that herbivory to a dominant plant species has community wide effects on pollination services. We observed a drastic change in the composition of pollinators visiting flowering plant species. Similarly, our results indicate that herbivory to the dominant plant species has negative effects on the reproduction of neighboring plants. We observed reductions in conspecific pollen deposition on three neighboring plant species. In V. cracca, reduced pollen deposition correlated with a notable reduction in seeds per fruit. These observations provide evidence corroborating the hypothesis that herbivory alters pollinator behavior and reduces plant reproduction at the community level. The reductions in pollination highlight the drastic role that herbivory can have in a community through indirect effects.
P.04 Pollinator diversity reduces pollination deficits through complementarity in thermal community niches
Matthias Albrecht Agroscope, Switzerland
The role of pollinator diversity in stabilizing crop pollination services and yield under climate change and increased weather variability remains poorly understood. Niche complementarity and increased response diversity of diverse crop pollinator communities resulting in broadened thermal community niches of crop pollinators could be particularly important to buffer pollination of early-flowering crops against high temperature variation in temperate production regions. We therefore investigated the impact of the thermal community niche of pollinators of early flowering sweet cherry relative to other potential explanatory variables in mitigating pollination deficits and on final crop yield across 48 intensively managed cherry orchards in Switzerland. A high variation in pollination deficits was observed across orchards, with an average deficit of 10% less fruit set of open-pollinated compared to supplementary hand-pollinated cherry flowers. Thermal community niche was the strongest predictor of the observed variation in pollination deficits and cherry yield among the studied potential drivers: enhanced community niche was associated with significantly reduced pollination deficits and increased final cherry yield. Our findings provide strong arguments for growers to invest in measures to maintain high levels of wild pollinator diversity on their farms.
P.05 Biodiversity of plants and pollinators in forested agricultural areas, can EUs agrilcultural policy affect it?
Georg Andersson, Yann Clough Centre for Environmental and Climate Research (CEC), Lund University, Sweden T
he European Common Agricultural policy, which is to be reformed soon, have been criticized for the low efficiency of reaching environmental goals. One of these is preserving biodiversity and also enhancing ecosystem functions, such as pollination. Our goal is to understand how agricultural support to farmers can affect the biodiversity in farmland in the forested areas in Sweden. This first step examined the biodiversity and abundance of pollinators depending on habitat, landscape type and forest cover in these areas.
P.06 Floral signalling variation, herbivory resistance and pollen limitation in the alpine rock-cress Arabis alpina
Sotiria Boutsi, Hampus Petrén, Magne Friberg Department of Biology, Lund University, Sweden
Interactions between plants and the surrounding community of pollinators and herbivores are central for shaping plant ecology and evolution. These interactions may contribute to intraspecific variation and local adaptation among different populations of a plant species. Increased knowledge of how local populations might vary in traits important for plant-insect interactions and how this variation is affected by various biotic and abiotic factors is important, not the least in alpine environments that are predicted to be especially vulnerable to anthropogenic change. In my master’s degree project, I use the alpine plant Arabis alpina (Brassicaceae), which is a model organism for both population genetics of alpine plants and the evolution of the selfing syndrome. However, so far only few studies have investigated the ecology and evolution of its biotic interactions. In my study, I will experimentally assess the presence of pollen limitation in 12 different self-incompatible Italian and Greek A. alpina populations. I will also test whether variation in pollen limitation among populations is associated with differences in floral signalling, herbivory resistance, population size or altitude, and identify potential evidence for inbreeding depression in small populations. Furthermore, I aim to investigate regional variation in floral signalling among these and potentially other populations, grown in the greenhouse, by analysing variation in floral scent, floral morphology, nectar composition and potentially plant defence traits. Given the drastic insect decline of the past decades, studies exploring the dependence of plants on their pollinators and the potential for local variation in such interactions, can provide valuable insight into the future of ecological communities.
P.07 Driven by mutualists? How declines in pollinators impact plant communities and ecosystem functioning
Yann Clough Centre for Environmental and Climate Research (CEC), Lund University, Sweden
Declines in pollinator availability – both in terms of abundance and species richness – causes concern not only for crop yields but also for the maintenance of pollination of wild plants, 80% of which are dependent on insects for pollination. In Europe, patches of semi-natural habitat embedded within a matrix of intensively used agricultural land may be particularly exposed, given the severe declines in insect biomass and diversity that have reported from such sites. Yet, we know surprisingly little about the effects of changes in pollinator availability on plant communities. Here, I first present an assessment of the effect of land-use intensity at the landscape scale on the degree to which plant communities are dominated by plants depend on pollinators. In that study we demonstrate consistent negative impacts of higher land-use intensity on the average dependence of plants on pollination (community-weighted mean pollinator dependence) in grasslands across Europe, when controlling for the local habitat quality of the focal grasslands showing that semi-natural grasslands in arable farming areas have less pollinator-dependent plants. Secondly, I discuss how the strength of this effect is likely to depend on pressure by herbivory and factors such as nutrient availability, using findings from common garden experiments and drawing upon a review of studies from theoretical and empirical ecology. Finally, I present a recently funded research programme, DrivenByPollinators, which aims at disentangling the effects of insects on plant communities and ecosystem functioning in landscapes of southern Sweden over the next five years.
P.08 Nocturnal bee pollination mediated by floral scents in Brazil
Guaraci D Cordeiro1, Cristiane Krug2, Mardiore Pinheiro3, Claudia I Silva4, Reisla Oliveira5, Irmgard Schäffler1, Theo Mota5, Clemens Schlindwein5, Isabel Alves dos Santos4, Stefan Dötterl1 1 University of Salzburg, Austria 2 Embrapa - Amazonia Ocidental, Manaus, Brazil 3 Universidade Federal da Fronteira Sul, Brazil 4 University of São Paulo, Brazil 5 Universidade Federal de Minas Gerais, Belo Horizonte, Brazil
Nocturnal pollination is mainly guided by floral scents. This is well documented for plants pollinated by bats, moths, and beetles, whereas the communication between nocturnal bees and their host plants remain poorly understood. We used a multidisciplinary approach to investigate the pollination and chemical communication between nocturnal bees and their host plants: cambuci (Campomanesia phaea, Myrtaceae), a fruit crop from the Atlantic Forest, jacarandá-do-cerrado (Macherium opacum, Fabaceae), a commom tree from Cerrado, and guaraná (Paullinia cupana, Sapindaceae), a fruit crop from the Amazon. We found that all three plant species depend on animal pollen vectors and open their flowers at night. Nocturnal bees of the genera Megalopta, Megommation, Ptiloglossa, and Zikanapis were the only effective (cambuci) or the most important (jacarandá-do-cerrado, guaraná) pollinators, with different bee species observed at the different plant species. The flowers of cambuci released 14 volatile compounds, mainly 2-phenylethanol, 1-octanol, 1-hexanol, and benzyl alcohol. In field bioassays at night, nocturnal bee pollinators were attracted by a synthetic scent blend of these compounds. Jacarandá-do-cerrado released 94 floral compounds, with the terpenoids α-copaene, α-terpineol, and β-myrcene being most abundant. Guaraná flowers emitted (E)-β-ocimene, linalool, and derivatives thereof (linalool oxides, lilac aldehyde, lilac alcohol) as most abundant compounds. A synthetic scent mixture resembling the scent of the flowers successfully attracted Megalopta bee pollinators at night. Our study highlights that nocturnal bees are the main pollinators of these plants, and that these bees, comparable to moths, bats, and nocturnal beetles, use floral scent cues to find their host plants. We also show that the scents of plants pollinated by nocturnal bees are highly variable and that the bees seem to be generalists that respond to various compounds.
P.09 Species and alien species
Tina D'Hertefeldt1, Helena Hanson2, Bente Eriksen, Cecilia Fredriksson3, Johanna Alkan Olsson2, Carina Sjöholm3, Niklas Vareman4 1 Department of Biology, Biodiversity Unite, Lund University, Sweden 2 Centre for Environmental and Climate Research (CEC), Lund University, Sweden 3 Department of Service Management and Service Studies, Lund University, Sweden 4 VBE, Lund University, Sweden
Through the recent attention on Invasive Alien Species and their effects on biodiversity, plants are now at the center of a new discourse in society. Trade and transport are known to be the main routes for introduction of invasive alien species and invasive species are one of the top-five threats to global biodiversity loss. At the same time, there is a need for introducing new plant species in a changing climate. This poster reports from a recently started project that aims to discuss the concept of invasive alien plants in a cross-disciplinary setting, with Sweden as a case study.
P.10 Wild bee and floral diversity co-vary in response to the direct and indirect impacts of land use
Christophe Dominik Helmholtz Centre for Environmental Research, Germany Wild pollinators and flowering plants in agricultural landscapes are threatened by habitat loss. While pollinators and insect-pollinated plants closely interact, it is still unclear how species richness and functional diversity of these two groups influence each other and how they respond to land use change. In this study, data from 24 agricultural landscapes in seven European countries were used to investigate the effect of landscape composition and habitat richness on species richness and functional diversity of wild bees and insect-pollinated plants. The relationships between the diversity of bees and flowering plants were characterized and indirect effects of landscape on bees and plants mediated by these relationships were identified. Increasing cover of arable land negatively affected flowering plant species richness, while increasing habitat richness positively affected the species richness and functional diversity of bees. In contrast, the functional diversity of insect-pollinated plants (when corrected for species richness) was unaffected by landscape composition, and habitat richness showed little relation to bee functional diversity. Additionally, bee species richness positively affected plant species richness and bee functional diversity were positively affected by both species richness and functional diversity of plants. The relationships between flowering plant and bee diversity were modulated by the indirect effects of landscape characteristics on the biotic communities. In conclusion, these findings demonstrate that landscape properties affect plant and bee communities in both direct and indirect ways. The interconnection between the diversities of wild bees and insect-pollinated plants increases the risk for parallel declines, extinctions, and functional depletion. This study highlights the necessity of considering the interplay between interacting species groups when assessing the response of entire communities to land use changes.
P.11 Strip-cropping of insect pollinated crops: Can crop diversity enhance vegetable production?
Natasha Holland, Yoko L. Dupont Aarhus University, Denmark
The demand for organic vegetables is increasing rapidly but cannot be met by current agronomic methods. Alternative methods, which enhance biodiversity and ecosystem services including pollination, are currently under development. In SureVeg (a Core Organic Co-fund funded project), we tested strip-cropping systems, i.e. a diversified cropping system in which alternating rows of different vegetable crops are sown. It is hypothesized that strip-cropping will provide floral resources during an extended period, supporting agroecosystem services provided by beneficial insects, including natural biological control and insect pollination. In the current study, we investigated a cropping system involving two insect pollinated crops: hokkaido pumpkin and faba beans, in addition to wild-flower strips. The study focused on pollination and fruit set of hokkaido pumpkin. The experimental set-up was established as an on-farm experiment hosted by an organic farm (Skiftekær Øko, Denmark), and included three fields: (1) hokkaido mono-culture (2) hokkaido mono-culture with flower strips and (3) strip cropped hokkaido pumpkin and faba bean and flower strips. In all fields, numbers of flowers and number and identity of flower-visiting insects were counted in transect walks carried out 2-3 times during the flowering season. Honeybees (Apis mellifera) were the main visitors, while bumblebees (Bombus spp) visited the flowers to a lesser extent. Pollinator abundances differed temporally and across different fields. Fruit weight was significantly higher in the strip cropped and flower strip hokkaido fields than the pure hokkaido field without flower strips. However, as the study did not include replicated fields, it cannot be concluded if other factors than cropping practice may have affected pollination and fruit set in hokkaido pumpkins. Furthermore, due to extreme drought in 2018, results may not reflect normal flowering and crop yield.
P.12 Integrating the management of pests and pollinators
Paul Egan1, Lynn Dicks2, Heikki Hokkanen3, Johan Stenberg1 1 Swedish University of Agricultural Sciences, Sweden 2 University of East Anglia, United Kingdom 3 University of Eastern Finland, Finland
To reduce the harmful impacts of pesticides on pollinators, policy makers and experts have increasingly turned to Integrated Pest Management (IPM). However, the extent to which non-pesticide IPM practices can also prove harmful to pollinators – and their ecosystem service – remains underappreciated. We articulate an urgent need to define more comprehensive approaches to IPM that show compatibility with crop pollination goals – but also vice versa. In this respect, we propose a first systematic framework for Integrated Pest and Pollinator Management (IPPM). Novel monitoring metrics (e.g. pollinator action thresholds) and a decision support tool (Pepo jEIL – ippm.shinyapps.io/server) are further introduced to operationalize IPPM.
P.13 The foraging behaviour of bumblebees in relation to landscape context
Janna Einöder Utrecht University, Netherlands Plant-pollinator interactions, among others between bumblebees and plants, are one of the most important mutualistic relationships globally maintaining both ecosystem functions and human food security. The mutualistic relationship between plants and bees might change when the environment changes, potentially leading to ecosystems disfunctions and food insecurity. To enable successful conservation measures, these often-complex interactions and relationships between species need to be understood better. In this study, we have therefore zoomed in on one of these crucial interactions: the foraging behaviour of bumblebees on wild plants in relation to landscape context and plant ecology. Bumblebees are central place foragers and are thought to follow an economically optimised way when foraging derived from the optimal foraging theory. Hence, bumblebees want to maximise high-reward visits while minimising their travel and search time as well as distance from the nest. By the aid of an event recorder software installed on a tablet, detailed behaviour of the bumblebees could be tracked resulting in over 70.000 observations from 800 bumblebees throughout the sampling sites. Tracked behaviour, for instance, was how much time they spent on each flower, plant or in the patch as well as if and how much pollen load they had and what they were foraging on (nectar/pollen). Additionally, it was studied if and under what circumstances bumblebees decided to change plant species, hence how strong their flower constancy was. The data is still being analysed, however, some trends for parts of the analyses become apparent: Optimal foraging theory is applied by some but not all bumblebee species and seems to be more influenced by the local scale than the landscape context. Hence, time in patches increased with patch quality but not with landscape quality. Daytime seem to affect foraging with bumblebees showing a higher time in patch in the afternoon than in the morning and increased wind speed decreased the bumblebees’ time in patch. We will continue to analyse the data to get deeper insights in the distinct behaviour of different bumblebee species as well as their flower constancy strategies.
P.14 The role of commercial ornamental plant varieties in supporting pollinator populations
Emily Erickson1, H M Patch1, S Adams2, L Russo3, V Wojcik4, C M Grozinger1 1 Pennsylvania State University, United States of America 2 Penn State Extension, United States of America 3 University of Tennessee, Knoxville, United States of America 4 Pollinator Partnership, United States of America
One of the major factors underpinning pollinator decline is the reduction in the diversity and abundance of flowering plant species. In urban and suburban areas, ornamental plants are commonly planted and could provide foraging resources for pollinator populations. However, their role in supporting pollinator biodiversity is not well established, and the few studies examining the attractiveness of these plants for pollinators have been conducted in urban areas which had a distinct pollinator community. In this study, we monitored pollinator visitation patterns to 50 ornamental annual and perennial plant cultivars over four years at two semi-natural sites in Pennsylvania, USA. We found that many cultivars vary in attractiveness based on time and year. Moreover, we observed only polylectic bee species visiting plants, despite the presence of oligolectic species in the background population. We conclude that the utility of ornamentals depends on environmental context: while their role in supporting a complex pollinator community is limited, they may provide long-lasting supplemental foraging resources in urban and suburban environments. Additionally, we found significant variation among cultivars in visitor abundance. Many of these cultivated varieties have a long history of artificial selection on floral traits such as color, shape, scent, and resource production, and this selection may have uncoupled traits that typically co-evolve to attract specific functional groups of pollinators. We are conducting choice assays using Bombus impatiens foragers to determine the relative influence of floral visual and chemical advertisement and nutritional reward on mediating patterns of pollinator attraction to these cultivated varieties. Overall, these studies (1) evaluate the potential of ornamental plants in supporting pollinator communities, (2) determine which plant features shape plant-pollinator interactions, and (3) assist growers to incorporate pollinator health into breeding and production practices.
P.15 Involving people to protect wild bees and other pollinators in the Mediterranean. Presentation of the project “LIFE 4 POLLINATORS”
Marta Galloni1, Laura Bortolotti2, Luis Navarro3, Theodora Petanidou4, Anna Traveset5 1 Department of Biological, Geological and Environmental Sciences, University of Bologna, Italy 2 Council for Agricultural Research and Economics, Bologna, Italy 3 Department of Plant Biology and Soil Sciences, University of Vigo, Vigo, Spain 4 Department of Geography, University of the Aegean, Mytilene, Greece 5 CSIC - Mediterranean Institute for Advanced Studies, Balearic Islands, Spain
Wild pollinators are the core of our ecosystems and their populations are declining dramatically in the last decades: land-use change, intensive agricultural management and pesticide use, all anthropogenic, are among the main causes. Out of the 1,965 species of wild bees reported in the European Red List of Bees, 9% are threatened with extinction, while 55.6% are indicated as “data deficient”: this lack of information regards in particular the Mediterranean basin (including Italy, Spain and Greece), which is considered as a “biodiversity hotspot” for conservation priorities. Public awareness on the role of wild pollinators needs to be improved especially in Mediterranean countries, where most of the existing awareness-raising initiatives still focus exclusively on honeybees. At the same time, agri-environmental measures have not been implemented at an adequate scale to compensate for the losses of suitable pollinator habitats. We present here the project: “Involving people to protect wild bees and other pollinators in the Mediterranean” (LIFE 4 POLLINATORS), recently submitted within the LIFE Governance and Information EU funding Programme. The main aim of the project is to gain environmental benefits by improving pollinator conservation, through the creation of a virtuous circuit that leads to a progressive change in practices across the Mediterranean region. In order to achieve this ambitious goal, the project has set a number of more specific objectives that foster education and dissemination, the implementation of citizen science approaches, and improved environmental governance. The partnership is built to cover a wide area of the Mediterranean region and its biological diversity, and to involve different actors ranging from the scientific perspective to key stakeholders, competent authorities and civil society.
P.16 Polyploidy, genetic leakage, reproductive success, and a geographic mosaic of specialized and generalized pollinators
Karin Gross1, Malin Undin2, Jodie Lilley3, John N Thompson4, Magne Friberg1 1 Department of Biology, Biodiversity Unit, Lund University, Sweden 2 Massey University, New Zealand 3 University of Manchester, United Kingdom 4 Department of Ecology and Evolutionary Biology, University of California, Santa Cruz, United States of America
Polyploidization, or whole-genome duplication, is an important mechanism of diversification in flowering plants. Even though it causes immediate strong reproductive isolation among different ploidy lineages, the isolation is often not complete. Polyploidy has also been shown to affect plant phenotype and the interaction with other organisms, such as pollinators. Here, we ask how hybrid crosses between different ploidy lineages affect reproductive success and floral trait diversification and how the novel variation is related to pollination by specialized and generalized pollinators. We assess the strength of reproductive isolation and the potential for genetic leakage among different ploidy types as well as the interaction dynamics with specialized and generalized pollinators in the woodland star (Lithophragma bolanderi). Recent studies have revealed that this species exhibits an incomparably high variation in floral traits and is composed of multiple ploidy types. It is pollinated by the highly specialized seed parasite Greya politella but also by generalized pollinators. We conducted a large-scale greenhouse crossing experiment within and between populations of different ploidy types. In addition, we assessed reproductive success in more than 20 natural populations across the entire distribution range of L. bolanderi. We present data on the seed set and germination rate of crosses within and between different ploidy lineages and the population pollination success in relation to the extent of pollination by the specialized pollinator G. politella and to plant ploidy-level. Such results provide the basis for our understanding of the mechanisms by which polyploidy affects diversification and how polyploid lineages can establish and coexist with their diploid progenitors.
P.17 An analysis of the effects of increased Carbon Dioxide on the protein and amino acid concentrations of pollen from Brassica rapa and Arabidopsis thaliana.
Emma Harrison, Perry Mitchell, Michael Taylor, Jennifer Roecklein-Canfield Simmons University, Boston, MA and Littleton High School, Littleton, MA
Colony Collapse Disorder (CCD) is a phenomenon in which worker bees disappear from the hive, leaving only the queen and immature bees. Incidents of CCD around the world pose an immense threat to global food security and economic growth. Theories regarding the cause of CCD include increased exposure to pesticides, pests and parasites, and habitat destruction. Additionally, poor nutrition is hypothesized to cause the death of worker bees associated with CCD. In this regard, some preliminary studies have shown that higher CO2 levels in the atmosphere, an imminent prediction of climate change, decrease the protein content in pollen and alter the essential amino acids that are crucial to bee nutrition. This implicates increased CO2 levels as a contributing factor to the malnutrition of bees and thus to the prevalence of CCD. To test this hypothesis, we grew Brassica rapa and Arabidopsis thaliana in both high and standard CO2 environments. We analyzed the protein and essential amino acid content of the pollen from both plants using HPLC and GCMS and we report our findings here. Preliminary studies showed a moderate decrease in protein in Brassica rapa and we undertake here a more in-depth study into how high CO2 affects the protein and amino acid content in pollen.
Bjørn Arild Hatteland NIBIO, Norway
P.19 Effect of air temperature on bumblebee behavior on drought stressed plants
Rebecca Julia Höfer, Manfred Ayasse, Jonas Kuppler Institute of Evolutionary Ecology and Conservation Genomics, Ulm University, Ulm, Germany
Climate change leads to increasing temperatures and reduced precipitation or extreme drought events. These can induce phenotypic alterations in plants and change flower visitors behavior, which in combination may alter trait-mediated flower-visitor interactions. In a wind tunnel experiment, we tested if different air temperatures modulate the behavior of Bombus terrestris in response to watered or drought stressed Sinapis arvensis plants. Further, we explored whether changes in behavior are linked to phenotypic changes in floral traits. We found that bumblebees did not preferred watered or drought stressed plants and the behavior did not differ between both groups. This can be explained by the similar floral trait expressions of watered and drought stressed plants. However, different temperatures resulted in behavioral changes of bumblebees such as finding the plants earlier at higher temperatures. In conclusion, our results indicate that plants can buffer floral trait expressions against short-term drought events to potentially maintain pollinator visitations, which were affected by air temperature. Therefore, simultaneously consider behavioral changes of pollinator and plant phenotypic expression in respond to temperature or precipitation can provide a more detailed picture how plant-pollinator interactions are altered by climate change.
P.20 Intraspecific variation of inflorescence characters: a plastic response to climate?
Gróa Valgerður Ingimundardóttir, Mikael Hedrén, Nils Cronberg, Torbjörn Tyler, Stefan Andersson Department of Biology, Lund University, Sweden
Autumn hawkbit, Scorzoneroides autumnalis, grows in a wide range of habitats in Europe, from the Pyrenees up to the northernmost reaches of Scandinavia. The species is highly variable and has several described varieties and ecotypes. In the Nordic countries, four varieties have been commonly recognized: var. pratensis, taraxaci, asperior and salinus. Some of the most important features used to distinguish between varieties have been the colour and hairiness of the involucral bracts, and the number of capitula. Plants with dark, hairy and few capitula (e.g. var. taraxaci) are particularly known from colder climate or alpine habitats whereas var. salinus, for example, is known from coastal meadows and is described to have hairless capitula. We have collected individuals of Scorzoneroides autumnalis from various habitats throughout Scandinavia and from Iceland, and grown them in a common garden in Lund, Sweden. In the common garden, we have over 70 populations representing several different coastal, alpine and ruderal habitats. Our preliminary field observations as well as studies of herbarium specimens from the Nordic countries, suggest that colour and hairiness of the involucral bracts, as well as the size and number of capitula, are environmentally plastic and appear to be related to trends in climate, with capitula being fewer, larger, hairier and darker in colder regions. Because dark coloration, hairiness and large flower-size are known to facilitate heat retention and insect visitation in other species, we hypothesize that S. autumnalis benefits from investing in these features under colder conditions. Studies are being planned to test this hypothesis.
P.21 Flora-wide trends in pollinator visitation – the role of plant life-history and mating system
Zdeněk Janovský Department of Botany, Faculty of Science, Charles University, Benátská 2, Prague 2, Czech Republic
Over the more than 150 years of research, a vast amount of data on pollinator visitation has been collected. Simultaneously, our knowledge of plant mating systems and generally life histories has greatly improved since the times of works of Knuth, MacLeod, Burkle and Willis. Do plants with particular life-history traits attract distinct pollinator assemblages? In order to answer this question, I compiled a database of pollinator observations for plants growing in Central and Northwest Europe from published literature. Since the sampling effort was not distributed homogeneously among the plant species, I included only species with at least 25 pollinator individuals recorded and reclassified their visitors into one of thirteen pollinator functional groups. Merging the pollinator database with the already existing databases on plant life histories and mating systems resulted in ca 200 species with known pollinator assemblages and life-history traits. Hoverflies and bumblebees were the two most abundant pollinator groups. Bumblebees on one hand hoverflies with muscid flies constituted the main gradient of variation in pollinator assemblages. Honeybees, butterflies and beetles dominated pollinator spectra of a small proportion of plants, but otherwise were rather infrequent. Solitary bees were rather thinly spread among the plant species, dominating pollinator assemblages of only few plant species. Hymenopterans frequented much more non-clonal perennials as compared to annuals and clonal perennials dominated by dipterans. Solitary bees and butterflies visited more often outcrossers especially those with some self-incompatibility mechanism. Beetles made up a larger proportion of pollinator assemblage only in frequently selfing plant species probably due to sparseness of visits by other pollinator groups. At a flora-wide level it seems, that the need of non-clonal perennials to invest more into outcrossing has led them to invest more into attracting more specific pollinators.
P.22 Floral evolution in two closely related orchid species: secondary evolution of pre-zygotic barriers?
Nina Joffard, Nina Sletvold EBC, Uppsala University, Sweden
When closely-related plant species coexist, competition for pollinators and selection against interspecific pollen transfer may promote pollination niche differentiation and floral divergence. Therefore, one may expect selection to maximize floral divergence between closely related plants in mixed compared to pure populations. Moreover, this process should be mirrored by a stronger phenotypic differentiation in mixed compared to pure populations (i.e. character displacement). In this project, we will compare flowering phenology, morphology and floral scent chemistry between two closely-related co-occurring orchid species in mixed versus pure populations to search for the signature of reinforcement.
P.23 The effect of pollination boxes, a means to increase densities of Eladobius kamerunicus, in an oil palm plantation in Indonesia
Lynn Jørgensen1, Anne Krag Brysting1, Douglas Sheil2, Anders Nielsen1 1 Centre for Ecological and Evolutionary Synthesis (CEES), Department of Biosciences, University of Oslo, Norway 2 Faculty of Environmental Sciences and Natural Resource Management, Norwegian University of Life Sciences, Campus Ås, Norway
My research was conducted in the oil palm plantation PT.Ketapang Agro Lestari (PT.KAL), in West Kalimantan, Indonesia.The oil palm (Elaeis guineensis) is a monoecious species, producing both male and female inflorescences at the same palm. The male inflorescence emits a strong anis-like scent during anthesis, attracting the pollinating African weevil Elaeidobius kamerunicus. E. kamerunicus feed on and deposit eggs inside male inflorescences in anthesis. During their visit to male flowers, weevils become covered with pollen grains which are subsequently transferred to female inflorescences that also emit an anis-scent, but do not produce any pollinator rewards. PT.KAL is testing a concept called “pollination boxes”, aiming to increase oil yields through increasing local weevil populations and thereby increasing pollination. Plantation workers manually collect and transfer post-anthesized male inflorescences to pollination boxes, each inflorescence potentially holding up to 3000 weevil eggs and larvae. The objective of my research is to examine the effect of pollination boxes on the relative abundance of weevils. I established transects at different locations within the plantation, each transect consisting of four sampling points located between 5-400m from the pollination boxes. I collected six random spikelets per inflorescence to assess the relative weevil abundance at each sampling point. Preliminary results suggest that distance to pollination boxes does not have a significant effect on relative weevil abundance. I also recorded the number of inflorescences, their sex, and their state of development in 24 palm trees surrounding each sampling point. I will further examine the effect of local inflorescence availability, and weather data, to evaluate the effect these variables may have on the relative abundance of weevils, and relate these data to yield data from the company.
P.24 Do bumblebees adapt to anthropogenic change in landscape and climate?
Cecilia Kardum Hjort Department of Biology, Lund University, Sweden
Since there has been a conceptual shift in our view of evolution, suggesting that a rapid adaptation to environmental change may be more common than previously thought, the aim of my project is to use bumblebees as a model organism to investigate the opportunities for genetic and morphological local adaptation due to landscape simplification and climate change.
P.25 Evaluating indirect effects of climate change on pollinators and pollination services through pesticide exposure
Jessica Knapp, Theresia Widhalm, Björn K Klatt, Ullrika Sahlin, Maj Rundlöf Lund University, Sweden, Post-doc, Oral presentation, Poster, Flash talk
Increased precipitation and a warmer climate are expected to intensify pest pressure and therefore, pesticide use. However, relatively little is known about the non-target effects of pesticides on bees and the pollination services that they provide. This is particularly true of the likely trade-off between the negative effects of pesticide exposure and the positive effects of improved flower quality. Therefore, across a gradient of landscape complexity, we aimed to quantify pesticide exposure in the pollen and nectar collected by honeybees and bumblebees (the risks), as well as the amount of pollen and nectar produced by our focal crop, red clover (the benefits). These findings will be related to the health and population success of sentinel bumblebee colonies at our study sites and, ultimately, to the pollination of red clover. With this data, we use existing bee models to predict population level effects of pesticide use and the likely impact that this will have on red clover seed yield. These findings will provide policy makers, land managers and farmers with a better understanding of the potential impacts that common management practices have, especially under predicted warmer climates.
P.26 Structures of flower surfaces in plants with contrasting pollination systems
Marjan Kraaij University of Groningen, The Netherlands
The surfaces of flowers come in different shapes and have different functions, but how they evolved remains largely unknown. Floral micro-texture can be a cue to insects, and increases in surface roughness by means of conical epidermal cells may facilitate flower handling by flower-landing insect pollinators. How epidermal cell shape and structure evolved with regard to the pollination system remains unknown. Here, we investigate the floral epidermal structure and shape in X species-pairs of Y families with contrasting pollination systems. We test whether flowers pollinated by (flower-landing) bees or flies feature more structured (rougher) surfaces than flowers pollinated by (non-landing) hawkmoths or birds and flowers that self-pollinate. In contrast to earlier studies, we find no evidence of co-evolution of flower surface and pollination system. The height, aspect ratio and overall roughness of floral epidermal cells varies between plant genera, but is not correlated with pollination system at large. Intriguingly however, we find that the upper (adaxial) flower surface that surrounds the reproductive organs and often constitutes the floral display is markedly more structured than the lower (abaxial) surface. We conclude that conical epidermal cells probably play a role in plant reproduction other than to provide grip to insects, such as hydrophobicity or improving the visual signal.
P.27 Can sexual selection cause divergence in mating system-related floral traits?
Åsa Lankinen1, Maria Strandh2 1 Plant Protection Biology, SLU Alnarp, Sweden 2 Department of Biology, Lund University, Sweden
The wide diversity of floral traits seen among plants is shaped by neutral and selective evolutionary processes. In outcrossing species, sexual selection from competing pollen donors is expected to be important for shaping mating system-related traits but empirical evidence is scarce. In a previous evaluation of experimental evolution lines crossed with either one or two pollen donors (monogamous, M, or polyandrous, P, lines) at early floral stages in mixed-mating Collinsia heterophylla (Plantaginaceae), P evolved enhanced pollen competitive ability and reduced maternal seed set compared to M, in accordance with sexually antagonistic evolution of pollen. Here, we asked whether the presence of sexual selection during pollen competition affects mating system-related floral traits in the same lines. We compared flowering start, timing of anther-stigma contact (as an indication of timing of self-pollination), timing of stigma receptivity and first seed set between M and P, and with a source line, S (starting material). The first three traits are later in outcrossers than in selfers of Collinsia. The last trait was expected to be earlier in P than in M resulting from sexual selection for early seed siring of pollen. Artificial polyandry for four generations resulted in later flowering start and later anther-stigma contact in P compared to M, and the latter trait was intermediate in S. Thus, P appeared more ‘outcrossing’ than M. Timing of stigma receptivity did not differ between lines. First seed set was earlier in P than in M, as expected from sexual selection. Our results from C. heterophylla experimental evolution lines suggest that a component of sexual selection during outcross pollination could enhance the patterns of floral divergence commonly found between outcrossers and selfers.
P.28 Why be equal? Ecological relationships between co-occurrent species with similar floral displays
Klaus-Rudolf Lunau1, Zong-Xin Ren2, Xiao-Qing Fan2, Graham H. Pyke3, Judith Trunschke2, Hong Wang2 1 Institute of Sensory Ecology, Heinrich-Heine-University Düsseldorf, Germany 2 Kunming Institute of Botany, CAS, Kunming, China 3 University of Technology Sydney, Australia
The visual signalling of pollen and anthers, and pollen- and anther-mimicking structures constitutes a very speciose mimicry system in flowering plants that are pollinated by pollen eating or pollen collecting animals like hoverflies and bees. By contrast, visual signalling of nectar and nectaries as well as false nectaries are rare among flowering plants. Here, we report on visual indication of nectar and nectaries found in a survey of flower colour diversity along an altitudinal gradient at the Yulong Snow Mountain in Yunnan (China). Using false colour photography we revealed small but conspicuous ultraviolet reflecting structures in the surrounding of the otherwise UV-absorbing central area of some flowers. The false colour photos are merged from a UV-photo and the blue and green channels of a colour photo while discarding the red channel, and thereby present the floral colour pattern as seen by UV-sensitive but red-insensitive bees. Visually conspicuous nectar and nectaries were found for example on the dark disc of Saxifraga melanocentra and Codonopsis graminifoli, the protuberances on the petals of Saxifraga wallichiana, and S. sinomontana, the staminodes of Parnassia wightiana, and Trollius yunnanensis, the unfolded petals Anaphalis nepalensis, the green floral guides of Solanum lyratum, and the black floral guides of Momordica cochinchinensis. The shared features of these structures are evident in the UV due to a glossy surface, the conspicuous colour contrast on the false colour photo, and the proximity to nectaries. We present evidence for the hypothesis that glistening surfaces on central areas of flowers indicate the presence of nectar and nectaries and function as visual nectar guides.
P.29 Effects of enemy community composition and drought on defense trait selection in woodland strawberry along a latitudinal gradient
Anne Muola1,2, Martijn L. Vandegehuchte3, Johan A. Stenberg2 1 Biodiversity Unit, University of Turku, Finland 2 Department of Plant Protection Biology, Swedish University of Agricultural Sciences, Sweden 3 Department of Biology, Ghent University, Belgium
Human-induced changes in the community complexity alter the strength of ecological interactions. Here, we focus on interactions between woodland strawberry (Fragaria vesca) and its natural enemies. As most plants, woodland strawberry has to defend itself against multiple enemies. In general, plants with fewer enemies are thought to invest less in defense and, therefore, more in growth and reproduction. However, several plant species have actually found to evolve higher defense levels if the enemy community becomes less complex. A novel, largely untested hypothesis suggests that multiple attackers often exert opposing selection on defense traits, preventing the plant from evolving successful defenses against multitude of enemies. Escape from complex enemy communities may enable the plant to adapt to selection imposed by fewer fitness-impacting species. The ability of a plant to efficiently defend itself is known to depend also on environmental stress. One of the major stressors resulting from climate change is summer droughts, which are predicted to increase both in frequency and severity. While our knowledge about the influence of enemy community complexity on plant defense evolution is limited, we understand even less how the effect of natural enemies on the ability of plants to evolve effective defenses interacts with abiotic stressors. To address these knowledge gaps, we are setting up a large-scale experiment along a latitudinal gradient in Europe stretching from the Mediterranean to the Northern Scandinavia. In each of the five common gardens we have 16 woodland strawberry genotypes from all across Europe in two treatments, control and drought.
P.30 Inferring gene flow in a spatially structured population of Primula farinosa
Etsuko Nonaka University of Jyväskylä, Finland
Landscape structure plays an important role in organizing biological diversity in landscapes by mediating evolutionary and ecological processes. Organisms interact with landscape structure to determine how they move about in their habitat to gather resources. As a result, such interactions can determine functional connectivity, the degree at which organisms connect populations and habitat elements in the landscape by movement and gene flow. Functional connectivity is crucial for determining spatial population and genetic structure of spatially structured populations. Using well-studied populations of an insect-pollinated, self-incompatible herb, Primula farinosa, as a concrete example, this project aims to theoretically and mechanistically link movement behavior of pollinators, the pattern of pollen flow, and landscape structure. It will utilize both genomic analysis and mathematical modeling approaches. We plan to genotype for dense SNP markers two generations of plants collected from multiple local population clusters found in Stora Alvaret in Öland, Sweden. Two-generation analyses will be conducted to assign a potential source population of pollen to offspring of sampled mothers whose genotype will be known. We also hope to quantify relative contribution of pollen and seeds to gene flow in the population using organelle DNA. Based on the information from these analyses, we will develop and parameterize mathematical models of pollinator movement to infer pollen flow between local plant populations in the landscape.
P.31 Quantitative pollen analysis using artificial intelligence
Ola Olsson Department of Biology, Biodiversity Unit, Lund University, Sweden We have developed a system for computer based, quantitative pollen analysis, using a Tensorflow based deep learning algorithm. The system relies on standard preparation of pollen samples in fuchsine gel on microscope slides, which are scanned at 0.25 μm resolution. We use a pollen library with over 200 plant species, which we have collected, prepared, and scanned. Individual pollen grains are labelled with species identity, and the algorithm is trained to identify these. The accuracy of the trained model is high, F1=90%, when identifying the same pollen images it was trained on. When identifying pollen in other samples accuracy is sometimes still quite impressive, but with some images much lower. We are currently (31 August, 2019) in a stage where we try to improve accuracy and reliability by various methods, and expect that soon (before 24 October, 2019…) we will have an efficient and trustworthy method, which can enhance efficiency of pollen analysis radically. Compiling a pollen library takes some time (but is probably needed regardless of method), and labelling pollen grains for training might take a few hours per species. Training the model as such usually takes less than two days. Once this is done, analysis of samples is quick and cheap: several pollen slides can be prepared per hour, scanning them takes 5-20 minutes per slide (of which only a few minutes need to be supervised), and then the AI analyses images in an automated manner; it counts and identifies ca 1500 pollen (or one typical sample) per hour (i.e. some 150-200 samples per week), and returns a tidy data file with the result, which are fully possible to cross-validate manually.
P.32 Using hierarchical joint models to study pollinator-mediated reproductive interactions
Øystein Opedal Research Centre for Ecological Change, University of Helsinki, Finland
Pollinator-mediated reproductive interactions among coflowering plant species are prime examples of how species interactions may affect fitness and community assembly. Despite considerable interest in these issues, statistical methods for assessing signal of reproductive interactions in observational data on coflowering species are currently lacking. I will present a flexible method for quantifying potential reproductive interactions among coflowering plant species using the hierarchical latent-variable joint models implemented in the Hierarchical Modelling of Species Communities (HMSC) framework. The method accommodates any measure of reproductive success, including pollinator visitation, stigma pollen loads, and seed set.
P.33 Phenotypic selection on floral traits in the perennial herb Arabis alpina
Hampus Petrén, Magne Friberg Department of Biology, Biodiversity Unit, Lund University, Sweden
Divergent natural selection may generate intraspecific variation in a species. More specifically, local selection patterns caused by abiotic and biotic factors, such as pollinators and herbivores, may vary among populations, generating inter-population differences in plant traits, such as flower size, floral scent and flowering phenology. Previously, we have documented extensive intraspecific variation in floral traits among European populations of the perennial alpine plant Arabis alpina (Brassicaceae). To investigate potential selection patterns responsible for some of this variation, we performed fieldwork in one Italian and one Greek A. alpina population during two years, measuring phenotypic selection on various floral traits, reproductive output, pollinator visitation rates, levels of herbivory and pollen limitation. Overall, we find some differences in the local pollinator communities and levels of pollen limitation, and limited divergence in plant traits and selection pressures. Using studies like this on phenotypic selection in natural populations, we can increase our understanding of when and how conditions in local populations may contribute to generating intraspecific variation among flowering plants.
P.34 Precision pollination of strawberries using chemical lures to attract syrphid species
Caroline Ponsonby University of Greenwich; Olombria, United Kingdom
Some 39% of insect visitations to globally important food crops are by insects other than bees and their importance is deeply understated in commercial pollination services. To better understand how to manipulate Syrphid (hoverfly) species in order to optimise their commercial pollinating potential, the effectiveness of semiochemical attractants in luring hoverflies to flowering soft fruit has been assessed. In vitro bioassays utilising Y-tube olfactometers, aerial arena and wind tunnel experiments were used to identify chemical compounds suitable for use in lures for semi-field trials. Results are still pending but early indications show great potential in their application for soft and top fruit pollination. Myophily can complement pre-existing bee pollinators and under certain circumstances, could replace them when used in conjunction with precision pollination monitoring systems designed by Olombria.
P.35 Effect of mass flowering crops on bumblebee communities in wild flower edges
Laura Riggi SLU Uppsala, Sweden Mass flowering crops can have a considerable effect on the flow of ecosystem services due to the large pulse of resources they provide. While the amount of floral resources is important for determining how many wild pollinators an area can sustain, an equally important – and relatively poorly investigated – factor is the timing of these resources. Bumblebees require a continuous, steady floral supply throughout the season to keep the colony running. Previous study have shown that periods of nectar surplus (i.e. spring bloom) are not enough to counteract the negative effects on bumblebees of low nectar resources later in the season (i.e. summer gap July - September). Late blooming mass flowering crops such as red clover (i.e. July – August) can be expected to benefit bumblebees and the services they provide. To test this we visited wild flower transects for bumblebees in landscapes with and without red clover crops during and after blooming in 2019. To test for pollination services at each site we measured fruit set in potted strawberries and beans. Current available data consists of over 3000-recorded bumblebee-flower interactions across 240 wild flower transects in Skåne. We aim to present preliminary results and hope to discuss hypotheses and methods during the poster session.
P.36 Effects of polyploidy and reproductive strategies on resource allocation and plant-insect interactions
Andrés Romero Bravo1, Karin Gross1, John N Thompson2, Kristina Karlsson Green3, Anna Runemark1, Magne Friberg1 1 Lund University, Sweden 2 University of California, Santa Cruz, USA 3 Swedish University of Agricultural Sciences, Sweden
Pollination, one of the most widespread form of plant-insect interaction, plays a central role in angiosperm and pollinator diversification by driving and shaping the existing genetic diversity. The most dramatic process generating this genetic diversity is polyploidization. Polyploidization is the duplication of the chromosome set and can result from one single genome (autopolyploidization) or from hybridization processes involving genomes of different taxa (allopolyploidization). Newly formed polyploid lineages (neopolyploids) usually face strong reproductive isolation from their diploid progenitors. The successful establishment of polyploid lineages is more likely to happen in lineages relying on asexual reproduction than in those conserving sexual reproduction. In this project, we investigate the evolutionary origin of polyploid lineages and the investment into different reproductive strategies in Lithophragma bolanderi, a plant species previously shown to consist of a geographic mosaic of several ploidy levels. We will perform a phylogenetic analysis on 29 different populations of L. bolanderi including also 9 other species in the genus to identify potential allopolyploidization events occurred along the evolutionary history of this taxon. In order to elucidate whether polyploid lineages of L. bolanderi depend on asexual reproduction, investing more resources in bulbil production than in sexual reproduction, reproductive strategies of different ploidy types will be characterized in a common garden experiment. We will compare bulbil weights from hand-pollinated plants (simulation of resource allocation into sexual reproduction) with those from a control group (simulation of resource allocation into asexual reproduction). Together, this combined approaches will help to disentangle the evolutionary history of the species and will provide new insights into how polyploidization events affect plant-insect interactions and coevolution.
P.37 Assessing the effect of domestic honey bees on wild pollinators in heathland using camera monitoring
Hjalte Ro-Poulsen University of Copenhagen, Denmark
The potential risk of mass-introduction of honey bees in natural and semi-natural habitats has concerned conservationist around the world. In Denmark, it has been heatedly debated during the past years, if competition between honeybees and wild bees in natural areas is a concern. Heathlands are among Special Areas of Conservation (SACs) and important habitat for flower-visiting insects during late summer, when the wide expanses of Calluna vulgaris are blooming. This further makes heathlands a popular destination for migratory beekeepers to set up hives. On Harrild heath, Denmark (1300 ha heathland) up to 450 hives are placed in different densities within the heathland during the flowering of C. vulgaris. In this study, we investigate the effect of mass introduction of commercial honey bees on wild pollinators (bees, hoverflies and butterflies). We use time-lapse camera-monitoring of C. vulgaris flowerbeds, in order to quantify visitation rates of honeybees and other flower-visiting insects throughout the flowering season. A total of 30 cameras were distributed around the heathland, encompassing locations in various distances from apiaries. Each camera was set to take a photo 40-45 cm distance from a flowerbed, every 60 seconds between 8h-18h. Using cameras to survey pollinators makes it possible to monitor many plots simultaneously and at a time-scale, which is not possible to do manually. This will provide new insight of the pollination dynamics and interactions of commercial honey bees and wild pollinators on heathlands.
P.38 Bees use fat to assess pollen quality
Fabian Ruedenauer1, Johannes Spaethe2, Sara Leonhardt1 1 Department of Zoology III, University of Würzburg, Würzburg, Germany 2 Department of Zoology II, University of Würzburg, Würzburg, Germany
Malnutrition is named a main reason for current bee population declines. Plant nectar and pollen are the only source of nutrients for almost all bee species. However, due to the variety of plants a bee has access to, it can be challenging to find the optimal diet. Several mechanisms could help bees to overcome this challenge, among them direct quality assessment at the flower, feedback on quality from the colony to the foragers or learned cues. Many species mainly use only one nutrient they regulate. We used chemotactile conditioning of the proboscis extension response (PER) to find out (1) whether bees can differentiate between different pollen species and (2) which nutrients they are able to taste and feeding assays to find out (3) which nutrients bees use to assess the pollen quality and how this influences their survival and reproduction and (4) whether the pollen they received as larvae is preferred due to learned cues preimaginally. Honeybees (Apis mellifera) and bumblebees (Bombus terrestris) were able to differentiate between pollen species. Additionally, bumblebees were able to perceive amino acids in water, but not in pollen, where instead they were able to perceive fatty acids. Accordingly, in the feeding assays they did not consume different amounts of pollen with different proportions of amino acids, but consumed less pollen with higher fatty acid contents, which decreased their survival, likely due to undereating other nutrients. Hence, bumblebees seem to regulate the fat content of pollen. Preimaginally learned cues, however, seem to play a subordinate role.
P.39 Why so different? The diversity and factors affecting nectar composition in populations of a rare plant Polemonium caeruleum L. in Poland
Justyna Ryniewicz1, Katarzyna Roguz1, Mateusz Skłodowski1, Andrzej Bajguz2 Marcin Zych1 1 Botanic Garden, Faculty of Biology, University of Warsaw, Warsaw, Poland 2 Department of Plant Biochemistry and Toxicology, Institute of Biology, Faculty of Biology and Chemistry, University of Bialystok, Bialystok, Poland
Polemonium caeruleum is a red-listed, boreal plant species with SW limit range in Poland. Flowers of this plant are visited by a broad spectrum of insects, indicating generalist pollination system. We chose 14 populations of P. caeruleum to our study, distributed across Poland. Experiments conducted in years 2014 to 2018 showed that populations differ in terms of seed production and frequency of insect visits. We speculated that nectar composition may be one of the factors involved in shaping those features. In 2018 we collected nectar samples from populations of P. caeruleum involved in our study. Sugars and amino acids (AAs) content was determined using high-performance liquid chromatography (HPLC). We found that nectar composition, regarding both sugars and AAs, is highly variable across populations. In contrast to previous studies, our findings demonstrate that the most common AAs and sugars in nectar of P. caeruleum were, respectively, glutamine and glutamic acid, and sucrose or fructose. Additionally, we have also specified some basic habitat parameters: habitat fertility (nitrogen, potassium and phosphorus content) and soil properties, for each of the population, that could explain the recorded differences. The project was supported financially by the Polish National Science Centre grant no. 2014/15/B/NZ8/00249 (to MZ).
P.40 Light at night – can it affect reproductive success of insect-pollinated plants?
Justyna Ryniewicz, Marcin Zych Botanic Garden, Faculty of Biology, University of Warsaw, Warsaw, Poland
Increasing light pollution during last decades seems to be a factor that significantly affects many groups of organisms and disturbs interactions between them. Also in the case of pollinators, the presence of artificial light at night can have a negative effect, influencing pollination services. In order to assess the impact of artificial night light on the reproductive success of insect-pollinated plants in 2019 we conducted a field experiment. We applied LED light, characterized by a broad spectrum of emitting wavelengths. As a research site we chose a meadow in the immediate vicinity of primeval Białowieża Forest, where the degree of light pollution is negligible. For the study we chose three plant species, characterized by different flower morphology and pollination biology: Aquilegia vulgaris pollinated predominantly by bumblebees active during the day, Saponaria officinalis moth-pollinated species, and Cosmos bipinnatus, a generalist species pollinated by a wide spectrum of insects. We set two experimental plots, where potted plants were exposed to pollinators: control site (natural day/light conditions), and experimental site with installed lightning at night. Within each site, we set a four-variant experiment to assess whether the presence of night LED light indirectly affects plant reproductive success (seed set) by influencing pollinators behavior. The project was supported financially by the University of Warsaw grant no. DSM 7600-33 (to JR).
P.41 Assessing market and non-market values of pollination services in Ireland
Jane C. Stout, James Murphy, Saorla Kavanagh Trinity College Dublin, Ireland
The Pollival project used pollinators and pollination services, which have public and political appeal, as a case study for assessing the market and non-market values of ecosystem services. To assess market values, we used agricultural food crop production and trade data (UN FAO) and published data on the degree of pollinator dependence for each crop. From these data, the global value of animal pollination to crop production was estimated at €158–412 billion. Using the same approach, the annual value of animal pollination to home-produced crops in Ireland was estimated to be €20–59 million per year. However, given the importance of international trade in animal-pollinated crops, and the fact that Ireland imports more than it produces of these crops, global animal pollinator decline could result in an increased trade deficit for these crops. Four economic scenarios predicted the cost of pollinator loss to Ireland at between €153 and €843 million per year. Thus, the risk of pollinator loss globally will have local market impacts in Ireland, in terms of increased food prices and an increased trade deficit in animal-pollinated crops. Scaling up to global level, this approach illustrates that pollinator loss can have differential impacts on national economies depending on the national balance of trade for animal-pollinated crops. In order to understand public perception of the importance of pollination services, and how they are valued by Irish society, we conducted national surveys. Of the 1000 randomly selected, representative respondents, the majority were aware that bees were in decline in Ireland, that it is important to protect bees and the benefits they provide, and that protecting the environment may require funding through taxation. On average, respondents indicated they were willing to pay an average of €4–6 per month (and up to €10) to protect bees and the flowers they pollinate, but further research will be required to develop a robust estimate of the willingness to pay for pollinator conservation. A second survey, carried out with a national newspaper showed also showed that more than half of respondents agreed that protecting pollinators may require funding through taxation, and most preferred the introduction of tariffs on products that harm pollinators and fines for actions that damage the places that pollinators live, breed or eat. Taken together, our results suggest that both the market and the non-market values of pollinators in Ireland are currently underestimated. There are many approaches to the valuation of ecosystem services, but market studies using analysis of global supply chains, and non-market approaches using methods such as willingness to pay, can reveal more about the monetary value of pollinators to the Irish economy. For a more holistic approach to assessing the values associated with pollination services, incorporating monetary and non-monetary approaches, a framework incorporating economic, social and health values of pollination services is required. By understanding and communicating the monetary and non-monetisable values of key ecosystem services, such as pollination, a better appreciation of natural capital can be developed for both policy and planning decisions at many levels across multiple sectors.
P.42 Does among-population variation in floral scent in Arabis alpina relate to phylogeographic history?
Kajsa Svensson, Hampus Petrén, Anna Runemark, Magne Friberg Department of Biology, Lund University, Sweden
Understanding the evolutionary processes behind population differentiation is an important topic in biology. In the perennial arctic-alpine herb Arabis alpina (Brassicaceae) previous research show that different European populations vary strikingly in the composition of floral scent, a trait known to play an important role in plant-insect interactions. In my study, I use genomic data from six A. alpina populations in Italy and Greece with varying scent profiles and try to identify if similarity in scent correspond to their phylogenetic relationship. In addition to identifying the phylogenetic relationship among these six populations, I also examine population differentiation and structure. My results show that all populations, except from the two Greek ones, are genetically differentiated from each other with varying magnitude, and that the differentiation rate was generally not related to similarity in scent. Furthermore, the phylogenetic and population structure analysis reveal that some populations with low similarity in scent are more closely related to each other than to other populations with a higher similarity in scent. Taken together, my results suggest that the spatial pattern of scent variation among these populations is not a result of phylogenetic relationship or genetic drift. Instead, the pattern may be a result of local adaptation, for example driven by plant-insect interactions, demonstrating how local selection in different populations may cause strong differentiation in floral scent phenotypes.
P.43 Insights from experimental hybridization on floral trait inheritance and reproductive isolation in self-compatible and self-incompatible Arabis alpina L. (Brassicaceae)
Hanna Thosteman Department of Biology, Lund University, Sweden
One of the main challenges in evolutionary ecology is to identify the factors and agents that drive diversification. Two of the most diverse groups of organisms on earth are plants and insects, and evidence strongly suggests that the diversification of both groups is driven by the interactions between them. One fruitful way for understanding how these interactions evolve and diversify is to identify systems of closely related species, or different intraspecific populations, where it is possible to investigate the same traits subject to different types of selection pressure. In the present study I measured flower size, anther orientation and herkogamy as well as floral scent emission rate and composition of alpine rock-cress Arabis alpina L. (Brassicaceae) to identify in what way they are affected by inbreeding and hybridization between self-compatible (SC) and self-incompatible (SI) populations. These populations differ from being completely depending on pollinator attraction for reproduction (SI-populations) to being almost autogamous (SC-populations with the ability to self-pollinate). A germination test revealed strong reproductive isolation between populations of different mating systems, with less than 10% of seeds produced in crosses between mating systems germinating, whereas 80-90% of seeds produced by crosses within-mating systems germinated. I found an effect of cross direction on flower size and floral scent composition, suggesting a potential presence of cytonuclear incompatibility between SC and SI populations. Furthermore, I discovered a loss of co-inheritance between flower size and total scent emission rate as individuals of different mating systems are crossed, suggesting that the two traits are not genetically connected. My findings constitute a foundation for future studies targeted to unravel the genetic background of floral trait variation in A. alpina. However, the discovery of strong reproductive isolation between SC and SI A. alpina indicate a potential need for reconsideration of the species classification and a necessity to split this species into two taxa.
P.44 Role of butterflies and hawkmoths in pollination networks on Mount Cameroon
Jan E.J. Mertens1, Lucas Brisson1,2, Yannick G. Klomberg1, Vincent Maicher1,3, Stepan Janecek1, Robert Tropek1,3, 1 Faculty of Science, Charles University, Prague, Czech Republic 2 Université de Poitiers, Poitiers Cedex, France 3 Biology Centre, Czech Academy of Science, České Budejovice, Czech Republic
Butterflies and hawkmoths are common flower visitors, but not much is known about environmental factors affecting their visitation rates and preferences. We studied how elevation and season influence role of butterflies and hawkmoths in pollination networks on Mount Cameroon, West Africa. We also tested which floral traits are crucial for butterflies and hawkmoths preferences. We video-recorded all flowering plants (>1200 plants of 217 species) at four elevations (650–2200 m asl), during both dry and wet season. All plants were video-recorded continuously for 24h and were checked afterwards for flower visitors. All butterflies and hawkmoths were identified to (morpho)species and their touches with reproductive organs and feeding behaviour were recorded. One third of the observed plant species across the whole gradient is visited by butterflies and/or hawkmoths, for a total of 743 individual flower visits by butterflies and hawkmoths. Butterfly visitation frequency and species richness decreased at higher elevations and during wet season. Hawkmoths seemed to play a more important role as pollinators at the higher elevations where butterfly visits were scarce (dry season) or absent (wet season). All networks were highly specialised with insignificant differences among elevations, seasons, and lepidopteran families. The pollination networks were more connected and less modular at higher elevations and during wet season. We also showed that butterflies and hawkmoths with longer proboscides preferred flowers with longer tubes. Similarly, larger species visited larger flowers. Hawkmoths and skippers surprisingly preferred similar flowers of large size and with long tubes, unlike other butterfly families. Only hawkmoths, skippers and papilionids preferred flowers rich for nectar sugars.
P.45 Pollination and plant reproduction above the tree line: the dominance of Diptera and increased selfing revisited
Judith Trunschke1, Zong-Xin Ren1, Klaus Lunau2, Graham Pyke3, Hong Wang1 1 Kunming Institute of Botany, Chinese Academy of Sciences, Kunming, China 2 Institute of Sensory Ecology, Heinrich-Heine-University, Düsseldorf, Germany 3 University of Technology Sydney, Ultimo, NSW, Australia; and Macquarie University, NSW, Australia
Plant-pollinator interactions in alpine habitats have been of great interest to both plant ecologists and entomologists because of the demanding environment characterizing living at high elevation. Since Müller´s (1881) first report on insect pollination of alpine plants a tremendous literature characterizes pollination in mountain habitats as being scarce and unpredictable, dominated by a fly pollinator community in line with often less conspicuous flowers, or alternatively, a high rate of self-pollination. Yet, this picture has largely emerged from a majority of studies conducted in the European Alps, Rocky Mountains and Andes while data from the Himalaya - home to the world´s highest peaks and a center of biodiversity - are largely missing. We explored the floral and pollinator species richness and functional diversity as well as plant reproductive success in three alpine plant communities in the eastern Himalayas in Yunnan, SW China, occurring at elevations of 3900 m, 4200 m and 4600 m a.s.l., respectively. In contrast to observations in other alpine regions, our data suggest that bumblebees and the native honeybee Apis cerana are the most important flower visitors while Diptera are nearly absent. Further, the majority of plant species show apparent floral morphological features that suggest a prevention of self-pollination, and thus, a strong dependence on these visitors for pollen transfer and seed production. Moreover, some of the species provide substantial amounts of pollen and nectar, which can support large bumblebee populations. In conclusion, we suggest that pollination by bees is increasingly important in these high mountain plant communities in the southeastern Himalaya. The unique environmental conditions of this region (formed by tree lines reaching beyond 4000 m a.s.l., high UV-radiation, summer rain seasonality) may explain this dominance of bumblebees compared to other mountain regions, but further research is needed to understand why flies are so rare.
P.46 What abiotic and biotic factors influence reproductive output the alpine species Ranunculus acris?
Linn Vasssvik1, Aud Halbritter2, Anne Krag Brysting1, Vigdis Vandvik2, Anders Nielsen1 1 Center for Ecological and Evolutionary Synthesis (CEES), Department of Bioscience, University of Oslo, Norway 2 Ecological and Environmental Change Research Group (EECRG), Department of Biology, University of Bergen, Norway
Reproductive output in alpine Ranunculus acris depend on abiotic and biotic factors. Alpine topography creates microhabitats for plants, where abiotic factors, like temperature and precipitation, play a big role in creating this heterogeneity. Snow cover, where it accumulates and the timing of snowmelt, are important drivers of the timing of life history events for plants. The initiation of flowering, the length of the growing season and the production of above ground biomass are all influenced by timing of snowmelt. These abiotic factors also affect the number of plant abundance and the number of pollinators available. This study investigates abiotic and biotic factors influencing reproductive output in alpine Ranunculus acris. I established 10 snowmelt gradients at Finse, southern alpine Norway, each gradient containing three stages (early, mid and late) representing three different timing of snowmelt. Different abiotic factors were measured along this gradient: timing of snowmelt and temperature. The total amount of R. acris plant biomass produced was weighed and abundance of R. acris individuals where counted throughout the growing season. In addition, a hand pollination experiment, to test for pollen limitation, was conducted. Seed mass (g) and seed:ovule ratio (n seeds produced / (n seeds produced + unfertilized ovules)) was used as measures of reproductive output. In the two years of this study the general trend was that a higher seed mass was produced in the early snowmelt stage, and decreasing in the later snowmelt stages, and that a higher plant biomass resulted in heavier seeds. However, in the second year of the study higher temperatures also resulted in higher seed mass, and a higher plant abundance in the surrounding vegetation had a decreasing effect on seed mass. Seed mass was not affected by pollen limitation. I found no relationship between seed:ovule ratio and any of the abiotic or biotic factors measured. My results show that R. acris growing in areas with an earlier snowmelt have a higher seed mass, which means more seeds are produced due to more time for fertilization and seed maturation. There is also less competition from the surrounding vegetation for nutrition and pollinators, which was also confirmed in the second year of the study, where higher plant abundance led to a lower seed mass. Biomass production is highly dependent on abiotic factors, like temperature, soil nutrition and soil moisture. Even though temperature only had a direct effect on seed mass in the second year of the study, indirect effects on seed mass from abiotic factors through the plant’s biomass could be just as important. This study highlights that alpine ecosystems are complex and that several factors, both biotic and abiotic, might have an important role in determining seed mass in R. acris.
P.47 The evolution of floral visual signals upon switches in pollination systems
Linda Verloop University of Groningen, The Netherlands T
he diversification of visual signals in flowers are largely driven by the interaction of flowers with pollinators. Pollinators (bees, birds, nocturnal moths and bats) see differently, and because flowers adjust their colors to the vision of their pollinators, visual signals of flowers differ depending on its pollinator. Pollinators of plants may shift during evolution and as a consequence the colors of flowers also shift. I study how coloration has evolved following transitions in pollination groups. I found evidence of convergent evolution of floral pigment absorbance spectra for pollination systems. Furthermore, I study how the optical principles of the petals’ light reflecting structures have evolved and to what extent this was driven by pollinators. In line with our hypotheses, preliminary results suggest that nocturnal flowers are more efficient and effective reflectors than diurnal flowers, that is, flowers pollinated at night have anatomies that enhance light reflection.
P.48 The importance of trees for bumblebees in semi-natural grasslands; explaining biodiversity, ecosystem services and rarity
Maarten Vervoort Lund University, Sweden
Changes in the agricultural landscape have caused declines in bumblebee diversity and abundance. Conservation efforts have been implemented, but it is necessary to expand our knowledge about the bumblebees’ ecology and its requirements to reverse the ongoing population declines. This study investigates the importance of trees within semi-natural grasslands to provide a high-quality habitat for bumblebees. Species richness and the abundance of bumblebees was measured in 34 semi-natural grasslands in southern Sweden. In addition, I quantified the presence of tree habitat in these grasslands and determined the relevance of food resources provided by trees. I found that the species richness of bumblebees is positively affected and that bumblebee abundance is negatively affected by the presence of trees. Furthermore, rare species increase in abundance and common species decrease in abundance as a result of increasing tree presence. Here, it is also shown that the provision of food resources by trees is explaining the significant effects on species richness and abundance of bumblebees. Based on my results, I suggest that trees can promote biodiversity but can also reduce ecosystem services. These findings also suggest that resource-providing trees are important for certain species bumblebees and therefore could explain rarity in bumblebees. Trees, when implemented correctly, improve habitat quality and benefit bumblebee presence in agricultural habitats.
P.49 Applying the natural capital management framework to wildflower meadows
Cian White, Jane Stout School of Natural Sciences, Trinity College Dublin, Ireland
Land managers are increasingly expected to manage landscapes for multiple benefits and purposes. Yet, despite proliferation of frameworks linking natural capital to ecosystem services, there remains little guidance for how management actions can improve ecosystem service provision. As ecosystem services cannot be directly manipulated, management actions must be targeted at natural capital stocks. Here I extend the Natural Capital Management Framework which explicitly links natural capital stocks to ecosystem service provision and identifies manageable natural capital stocks as the critical intervention point. Wildflower meadows and three services they deliver are used to demonstrate the utility of the framework. When managing for ecosystem services it is useful to consider ecosystem service providers; the set of species which contribute to the supply of all services. The ecosystem service provider set can be divided into subsets based on response traits, traits which determine how a species responds to a management intervention. Grouping the ecosystem service providers by response traits produces species cohorts which respond similarly to a management intervention and can be considered a manageable natural capital stock. A natural capital stock has three dimensions in which it can be manipulated: quantity, quality and spatial configuration. Manipulating one dimension of a stock reveals how it modulates the flow of ecosystem services, describing a dimension – ecosystem service supply relationship. Elucidating these relationships is an important task in the context of designing management strategies efficiently. I consider the floral community of wildflower meadows as a natural capital stock grouped by response traits, and investigate how two dimensions of the stock, quantity and quality, modulate the supply of three ecosystem services, preservation of rare species, pollination and psychological well-being.
P.50 Measuring pollination services in red clover fields
Theresia Widhalm, Jessica Knapp, Maj Rundlöf Lund University, Sweden
Pollination services are of crucial importance for biodiversity conservation as well as agricultural production because a majority of flowering plants depend on animals for reproduction. However, multiple environmental stressors are currently affecting pollinators, and declines have been recorded throughout the world. The resulting potential lack in pollination services can be quantified by measuring the difference between optimal and current levels of pollination; the pollination deficit. We determined the pollinator dependency and pollination deficit in conventionally and organically managed red clover (Trifolium pratense) cultivated for seed production. Due to the large number of clover cultivars grown by farmers, we measured each cultivars level of pollinator dependence as the amount of yield under animal pollinator exclosure conditions. To quantify the pollination deficit, we implemented open pollination and maximised pollination as treatment levels; the latter being a combination of open pollination and hand pollination. Furthermore, as hand-pollinating clover under field conditions is a novel approach, we measured the effectiveness of our method by controlling for potential effects of parts of the hand pollination process not directly linked to pollen transfer. We used three additional pollination treatments (wing petal removal only, wing petal removal under animal pollinator exclosure, hand pollination under animal pollinator exclosure) as experimental controls of the maximised pollination treatment described above. Here, we report on the methodology for quantifying pollinator dependence and pollination deficit in red clover under field conditions. Our methods can be used to better understand how impacts of different management practices affect pollination and seed production in red clover.
P.51 Population dynamics of the butterfly Pyrgus armoricanus after translocation beyond its northern range margin
Theresia Widhalm1, Yoan Fourcade2, Thomas Frank3, Erik Öckinger2 1 Lund University, Sweden 2 SLU Uppsala, Sweden 3 BOKU Vienna, Austria
Translocation experiments can be used to study the factors limiting species’ distributions and to infer potential drivers of successful colonisation during range shifts. To study the expansion dynamics of the butterfly Pyrgus armoricanus in southern Sweden and to find out whether its distribution was limited by climate, translocation experiments were carried out within and 50-60 km beyond its natural range margin. Populations were monitored for eight years following the translocation. Although most translocation attempts failed, P. armoricanus was able to survive in two sites north of its current range limit. One of them eventually led to expansion and establishment of a viable metapopulation. Translocation success appeared to be independent of latitude, suggesting that climate is not the main factor determining the current northern distribution limits of this butterfly. Population growth and secondary spread in the expanding population was positively related to patch area, connectivity and negatively related to barriers to dispersal. Local habitat quality seemed to be less important, although microclimate may play a role in colonisation success. The limited success of the translocation and the importance of a well-connected patch network suggest that the current distribution of P. armoricanus is limited by its low dispersal ability combined with the fragmentation of its habitat, making it unlikely to track its changing climatic niche. Assisted migration could be an effective tool for such species, but long-term evidence for its effectiveness is not yet available.
P.52 Trees a better predictor for solitary bee fitness than semi-natural habitat- and oilseed rape availability
Johanna Yourstone, Melanie Karlsson, Melanie Karlsson, Björn Klatt, Ola Olsson, Henrik Smith Lund University, Sweden
Declines in wild pollinators, such as bees, constitute a threat to both wild plants and crop production. To be able to mitigate bee declines it is important to understand how bees respond to e.g. a lower availability of wild flowers growing in semi-natural habitats, and a varying availability of extremely resource-rich, but ephemeral, mass-flowering crops in modern simplified agricultural landscapes. In our study we focused on the cavity-nesting solitary bee Osmia bicornis. We investigated if and at what spatial scale the species is benefitted by the mass-flowering crop oilseed rape, as well as by the availability of semi-natural habitats and trees in a modern agricultural landscape. We also aimed to investigate the diet of the bee in relation to available resources. Our study was set in 12 areas that varied in their extent of semi-natural habitats and tree abundance in southern Sweden. In each area we placed trap nests loaded with cocoons of O. bicornis at about 0, 300 and 1000 m distance from nearest oilseed rape field. We sampled pollen from the bee nests and assessed the bees’ reproduction success. We also made detailed surveys of semi-natural habitats and trees within 500 m from each nest. We found that the amount of semi-natural habitats was not related to the reproduction success of O. bicornis, while both oilseed rape and the amount of trees were positively related to the reproduction. The pollen samples showed that O. bicornis mainly collected oak and other tree species pollen in the beginning of the nesting season, turning to mainly buttercup and other grassland species later on. Our results highlight the hitherto overlooked importance of trees to wild bees, which deserves further investigation in order to plan accurate conservation actions for bees in modern agricultural landscapes.
P.53 Untangling temporal and spatial dynamics in a plant-butterfly ecological network
Konstantina Zografou1,2, Mark T Swartz3, Virginia P Tilden1,3, Erika N McKinney1,3, Julie A Eckenrode1,3, Brent J Sewall1 1 Department of Biology, Temple University, Philadelphia, Pennsylvania, United States of America 2 Institute for Ecology and Evolution, University of Bern, Bern, Switzerland 3 The Pennsylvania Department of Military and Veterans Affairs, Fort Indiantown Gap National Guard Training Center, Annville, United States of America
A framework to study plant-animal mutualisms is ecological networks. Although studies on networks have returned important insights into the complex structure of natural communities, relatively little is known about the temporal dynamics of species interaction networks. This obscures our understanding of how natural communities respond to global environmental change. To come closer to a more comprehensive understanding of world we studied the temporal and spatial dynamics of a butterfly-plant network (2007-2017). Across years, we found network structure (connectance, specialization, nestedness) to remain consistent. In contrast, we revealed strong instability for temporal dynamics of species and their interactions together with a bimodal variation in temporal persistence for butterflies. Bimodality corresponded to sporadic and stable species that differ in their level of persistence in the network. Topological maps of species temporal persistence and linkage level revealed the existence of two distinct regions of “interest” as sporadic-specialists and stable-generalists and signs for a third one, largely occupied by invasive plant species. A unique pattern of increasing butterflies, plants and interactions in July-August (middle months) and connectance and specialization index in June-September (border months) was assessed at the finest grade temporal scale, among seasons. Spatially, only one site was found to differ significantly from the rest of the study system. Incorporating seasonal variation in our analytical framework, it has greatly improved our understanding on the network structuring yielding September, as the month with the highest connectivity and network specialization. Our study has consistent evidence that mutualistic webs are highly dynamic entities, but also exposes the source of temporal variability that if taken into consideration, will greatly improve our ability to understand of how nature evolves to ongoing changes.