Plant-pollinator networks in a restoration planting, and the effects of non-native plants and nitrogen fertilisation

Bridget Anne Johnson

    Research output: ThesisDoctoral Thesis

    557 Downloads (Pure)


    Global changes can act individually or in combination to change plant-pollinator interactions. Impacts of global changes can influence both plant reproductive traits and/or pollinator behaviour. This PhD study aims to further understand how plant- pollinator interactions are influenced by non-native plant invasion and addition of nitrogen (N) in the context of a revegetated agroecosystem. The study was completed within a large restoration experiment in agricultural land in southwestern Australia. My research examined the following overarching questions:
    1. Does the efficiency of pollinators change in the presence of non-native flowering plants and the addition of N fertilisation?
    2. How does the flowering plant community composition and N fertilisation influence plant-pollinator network structure?
    3. Does N fertilisation change the composition of nectar traits and visitation rates of the European honey bee (Apis mellifera)?

    First, I investigated how the presence of non-native plants and addition of nitrogen changes the relative importance of floral visitors as pollinators of four native plant species. I observed and captured floral visitors and calculated the Pollination Importance Value (relative effectiveness) of each insect morphospecies in the presence or absence of non-native plants, and with or without the addition of N. I observed 8936 floral visitors of 249 morphospecies from 14 insect orders, on four native woody species planted into the site (Banksia sessilis, Hakea lissocarpha, Callistemon phoeniceus, and Calothamnus quadrifidus) and two non-native herbaceous species present from the previous pasture community (Arctotheca calendula and Echium plantagineum). The non-native A. mellifera was the most dominant and effective pollinator of all four native plant species. Experimental exclusion of floral visitors revealed that insect pollinators were required for the seed set of three of the native species, and that the native plants at the study site were not pollen limited. The two treatments did not affect the effectiveness of the potential pollinators, or the proportion of viable seeds produced in the four native plants.

    After identifying the potential pollinators at the field site, I determined how thestructure of plant-pollinator networks changes with the presence of non-native plants and N deposition. I observed insects visiting flowers in plant assemblages of four ratios of native: non-native flowering plant species, combined with a treatment of N addition. I used linear mixed models (GLMM) to determine the most important factors affecting relative pollinator abundance and morphospecies richness. To determine the structure of each plant assemblage ratio, I used network analysis and compared four common network indices to observed values with null models. Our results illustrated that plant assemblages with higher native-plant richness received a high number of pollinator visits and attracted greater morphospecies richness. The addition of N influenced the network index plant generality. I conclude that established and functioning plant-pollinator networks can develop in recently revegetated sites comprising a mix of native and non-native flowering plants.

    I explored the effects of N fertilisation on nectar traits and visitation rates of the most common pollinator at the field site, A. mellifera. I collected nectar from and observed A. mellifera visitation of two native flowering plants, C. phoeniceus and C. quadrifidus, with and without N fertilisation and herbicide. Nectar samples were analysed for the concentration of three sugars (glucose, fructose and sucrose) and 20 amino acids using high-performance liquid chromatography. I used GLMMs to analyse how nitrogen and herbicide affected six response variables. Both native plants were dominated by the proline. In the native plants, N fertilisation increased the amount of glucose, changes amino acid composition and the concentration of four amino acids (histidine/glutamine, proline, valine, and lysine) essential for honey bee survival, and increases the time a honey bee spends on flowers.

    In summary, this thesis describes a previously unknown pollinator community and presents an understanding of how plant-pollinator interactions develop in an assembling community following restoration, and how these are influenced by the presence of flowering non-native plants and addition of N fertilisation. This study highlights that mixed assemblages of native and non-native flowering species can support established plant-pollinator networks. The results suggest that it might not be necessary to remove non-native species for restoration of plant-pollinator networks in early stages of restoration. This research provides a foundation for further analyses of the development of plant-pollinator networks as the restoration progresses.
    Original languageEnglish
    QualificationDoctor of Philosophy
    Awarding Institution
    • The University of Western Australia
    Award date8 Jul 2016
    Publication statusUnpublished - 2015


    Dive into the research topics of 'Plant-pollinator networks in a restoration planting, and the effects of non-native plants and nitrogen fertilisation'. Together they form a unique fingerprint.

    Cite this