Plant-pollinator interactions represent one of the most important mutualisms in nature. Multiple complex floral characters function as pollinator attractors, including flower size, colour, shape as well as nectar and pollen. The diversity and abundance of flowering plant species and associated floral rewards drives pollinator population abundance and community diversity. Within agroecosystems, large-scale, mass-flowering monocultures associated with many crop species offer wild pollinators only a short-term bountiful floral resource in an otherwise florally depauperate landscape. Outside this window, resources may be scarce, with important implications for wild pollinator populations and consequently pollination services to native and crop plants, as well as food security. We demonstrate the role that targeted plantings of locally adapted, native floral resources can play in supporting native pollinator populations within agroecosystems by filling temporal and spatial gaps in the diversity and availability of resources. Our study assessed flowering phenology and pollinator visitation for a range of floral species native to the Sydney Region. We identified 41 native plant species that successively flower throughout the year, providing continuity of diverse floral resources year-round that attracted a diverse array of native pollinator groups. A total of 18 native insect species were recorded as flower visitors, with different seasonal patterns in abundance observed. Preliminary analysis indicates that flower numbers and floral traits including colour and nectar availability were drivers of pollinator diversity and abundance. Selected plant species will be established as on-farm floral enhancements in apple orchards to support wild insect pollinators, thus reducing agricultural dependence on introduced, managed honeybees.

Objectives: The detection and monitoring of changes in plant community composition is becoming an increasing concern, as climate change begins to impact the occurrence and distribution of many plant species. Plant identification from pollen carried by pollinators is an approach that could be employed to monitor vegetation changes in a landscape. Pollen DNA metabarcoding, a novel genomic method, has been recently demonstrated to improve detection and identification of plant species from a mixed pollen sample compared to traditional microscopy. We applied pollen DNA metabarcoding in the identification of plant species from pollen collected by European honeybees (Apis mellifera), a widely-distributed generalist pollinator, from managed beehives located within an urban reserve.

Methods: Using two metabarcoding markers, trnL and ITS2, we analysed the plant composition of three different pollen sources associated with honeybees: pollen from the bodies of individual bees, pollen collected in pollen traps and honey produced by the hive.

Results: Pollen metabarcoding from individual bees was able to detect up to 50% of the species observed flowering in the study area, but also many other species initially undetected by our vegetation survey. We also compared how each of the pollen sources (individual bees, honey and pollen traps) reflects the composition of the surrounding plant community, with each substrate representing a different time window or level of resolution.

Conclusions: We discuss the potential of using honeybees as biomonitors of plant diversity, and the possible role of pollen DNA analysis using managed beehives in the long-term monitoring of plant community structure.

Pollinators are currently facing dramatic declines in abundance and richness. This can have profound impacts on agriculture, as 75% of crops benefit from pollination services. Native bees depend on natural areas for nesting and flowering sources; therefore, restoring natural areas within croplands can be beneficial for biodiversity and economic development. Despite the advantages, restoration faces implementation challenges, as it typically requires upfront costs and taking some land out of production. Designing sustainable landscapes in this context will require novel approaches, but there is limited knowledge and tools for how to include the potential of pollination services in spatial restoration planning. Here, we develop a systematic planning framework to determine the best locations for restoration that account for the improved yields delivered by pollination services. We demonstrate the potential of this framework to improve land-use decisions in a coffee production landscape in Costa Rica across a range of scenarios. Our work quantifies the extent to which restoration can provide synergies between agricultural production and conservation. Our results illustrate that restoration can provide economic benefits in the long run, which can serve as a tool for engaging local land-holders in conservation endeavours which also enhance their livelihoods.

Biodiversity influences ecosystem function, but there is limited understanding of the mechanisms that support this relationship across different land use types in mosaic agroecosystems. Network approaches can help to understand how community structure influences ecosystem function across landscapes. However, in ecology, network analyses have largely focused on species–species interactions. We test the utility of bipartite network analysis to identify conservation priorities to support ecosystem services in agricultural landscapes. We use this method to link insect pollinator communities to sites in a tropical agricultural landscape and examine how the structure of the community network influences plant reproduction in Brassica rapa. Our approach showed that Diptera: Syrphidae visits were the strongest contributor to the number of fertilized pods, while visits by Syrphidae, Hymenoptera and Lepidoptera had the strongest effect on the number of seeds per pod. Sentinel pots at forest sites were visited by more unique species (i.e. species with higher specialisation, d’) than sites in other land uses, and dairy sites had more visitors that were common across the network. Participation coefficients, which indicate how connected a single node is across network modules, were strong predictors of ecosystem function: plant reproduction increased at sites with higher participation coefficients. Previous studies have shown that participation coefficients influence function in metabolic and neurological networks; here we have shown a relationship between participation coefficients and ecological function. Identifying ‘keystone’ species and sites that have a strong influence on network structure can inform conservation priorities and decision-making in diverse agroecosystems.

Insectivorous bats provide an important ecosystem service by controlling the distribution and population sizes of a large number of prey species. In agricultural areas, pest control services provided by insectivorous bats are economically indispensable. Understanding the ecology and habitat requirements of insectivorous bats is central to their conservation and it is critical to develop practical methods to understand their ecological requirements. Stable isotope analysis is known to be a practical tool to answer trophic questions about insectivorous bats, and to overcome methodological problems associated with conventional dietary studies.

This study elucidated foraging patterns of two insectivorous bat species (Vespadelus vulturnus and Nyctophilus geoffroyi) in central western New South Wales, by analysing the isotopic signatures (δ13C and δ15N) of the dominant prey species from two families of moths (Noctuidae and Aganaidae), and one family of beetle (Elateridae).

The isotopic differences of hair and faecal samples from V. vulturnus and N. geoffroyi revealed seasonal shifts in diet using general linear models and a multiple-source mixing model. The results indicated that V. vulturnus incorporated a higher proportion of beetles throughout both the summer and autumn months, whereas N. geoffroyi incorporated a higher proportion of beetles in their diet during the summer months, and a higher proportion of moths in their diets during the autumn months.

The results from this study showed that stable isotope analysis was a practical tool in understanding the trophic interactions of insectivorous bats, and can offer an alternative method in this field where conventional methods may fall short.

Whitehaven Coal’s use of ecological burns in Biodiversity Offsets for the Maules Creek Coal Mine is demonstrating leading practice for the dual purpose of ecological restoration and bushfire hazard mitigation. Detailed statistical analysis of the annual vegetation monitoring results has found a clear positive effect from the introduction of fire, resulting in an increase in both native species richness and frequency. Whitehaven Coal has practicably demonstrated that the use of fire can progress the Biodiversity Offsets towards Completion Criteria targets. The use of experienced professional hazard reduction fire contractors means the ecological burns are implemented competently and safely with the support of key regulators and the community. Whitehaven Coal began trialing ecological burns within Biodiversity Offsets back in 2013; these results endorse the change to use ecological burns as an annual routine biodiversity management strategy since 2017.
The results for the ecological burn at the Wirradale Offset are clear – the use of fire resulted in a positive change in vegetation condition, despite a relatively short period of time post fire and less than 12 months between monitoring. After discounting for the influence of other treatments and effects (like rainfall); the statistical analysis found:
• Strong evidence (equates to a p-value of 0.05-0.005) that burning alone increases native richness (by ~1-2 species) across the burn treatment sites compared to unburnt and other treatment sites;
• Burning appears to be beneficial for native species richness; and
• Burning has a positive effect on both occurrence and frequency of native species.

Future land-use change could supplement populations of opportunistic predatory birds, such as corvids, resulting in amplified predation pressure and negative effects on populations of other avian species. We investigated whether Forest Ravens (Corvus tasmanicus) were more likely to be observed in modified landscapes and in areas of higher roadkill density in south-eastern Tasmania. Following this, we examined the effect of Forest Raven density on the abundance of other birds. We surveyed birds along roadsides to investigate the effects of land-use and raven population density on the presence of smaller birds/songbirds. We used species distribution models and generalised linear models to assess the habitat and population dynamics of Forest Ravens. Roadkill presence increased the probability of observing ravens more than six-fold. Forest Raven presence was also higher in agricultural areas than in forested and urban areas. There was no effect of Forest Raven abundance on species richness or abundance of smaller songbirds. This null result may be due to seasonal effects or because ravens moderate the presence or abundance of other avian predators. However, the interactions of meso-predators with other species are complex, and further work is required to assess the long-term effects of anthropogenic supplementation of ravens on the ecosystem.

Many bird species are declining globally in rural landscapes through habitat loss and modification associated with changing land-use. Identifying landscape attributes that are important for maintaining diverse communities is crucial for conservation. I quantified landscape attributes in 25 study landscapes, (each 1 km²), in the Strzelecki Ranges, Victoria, including: habitat extent (native wooded cover, exotic wooded, open pasture); composition (elevation range, scattered tree density, average residential area), spatial configuration (aggregation of native vegetation), and context (distance to forest ‘source’ area >50 ha). Point-counts to survey birds were undertaken at 10 sites per landscape, and data were compiled for four groups of species: forest, open tolerant, open country and exotic species. Multivariate modelling compared the influence of landscape attributes on each bird group in ‘all landscapes’ (n = 25 landscapes, range of 3 – 99% native wooded cover), and a subset of ‘modified landscapes’ (n = 19, 3 – 29%). Forest species richness increased in landscapes with greater native wooded cover and farmland exotic vegetation associated with nearby forest. Seven of 46 forest species were absent from the modified landscapes. The attributes did not strongly affect the richness of open tolerant and open country species in the modified landscapes. Exotic species richness responded positively to residential property area and negatively to native forest cover. Farmland mosaics (e.g. wooded habitats dispersed amongst areas of pasture) and large source areas of intact forest are important for rural bird communities. Future conservation needs to protect these habitats for biodiversity as rural landscape change continues.