The project had two parallel elements: 1) a field experiment where we exposed common, native perennial plants to runoff concentrations of agrochemicals, and 2) a glasshouse study where we grew the same species of plants in pots, and exposed them to the same treatments, in order to collect pollen and nectar. We repeated the field experiment for two years, and thus had 8 replicates of the treatment effect in the first year and 4 of the long term treatment effect in the second year. We also conducted an analysis to determine whether soil quality 1) was affected by our treatments and 2) affected plant health and pollinator visitation.
We installed the first 16 plots (4 replicates of 4 treatments) in 2017, working with businesses in Dublin interested in biodiversity initiatives. The experimental treatments were: 1) runoff fertiliser, 2) runoff herbicide, 3) combination of runoff fertiliser and herbicide, and 4) water control. We monitored plant health and watched each flowering plant species in each plot for 5 minutes at a time, and collecting any flower-visiting insects with an insect vacuum throughout the summer. In 2018, 16 new plots at 4 new sites were sown in the same way as the 2017 plots, and treated and monitored for plant health and pollinators throughout the year. We conducted a destructive survey at the end of 2018, measuring the height of flowering plants and longest leaf length of non-flowering plants, as well as the proportion of flowering plants. For plots in their second year of growth, we counted the number of flowering and non-flowering stems and measured the height of flowering stems and the leaf length of non-flowering plants.
In the glasshouse, we were able to collect sufficient pollen from 6 of the 7, and nectar from 3 of 7 of the plant species. We collaborated with three European universities to analyse these samples for amino acids, fatty acids, and sugars. We have data from two of the three analyses and this work is ongoing.
We also analysed of the effect of habitat types on plant-pollinator networks, using the largest dataset of plant-pollinator interactions ever collected in Ireland. This study also allowed us to provide evidence-based recommendations for pollinator plantings in Ireland. We then compared the Irish plant-pollinator networks to other publicly available plant-pollinator networks to determine whether the Irish networks differed.
Results so far:
1) The treatments had significant effects on plant health, inducing changes in the timing and probability of flowering, the size of the floral display, and the height and leaf length of plants.
2) Plant health affects pollinators in two ways: 1) it affects the floral display with a direct effect on pollinator visitation, and 2) there is also an indirect effect of plant health in pollinator preference, suggesting some change in the quality of the floral resources or the attractiveness of the plants.
3) Our analysis shows no significant difference in the soil quality in plots before and after the treatments were applied. Significant differences between soil quality at different research sites did help to explain variation in plant health between sites.
4) There was no difference in the quantity of floral resources (nectar/pollen) between treatments.
5) There are significant effects of experimental treatment on nutritional attributes of the nectar and pollen.
6) Irish plant-pollinator networks are affected by habitat.
7) Irish plant-pollinator networks differ from other networks around the globe. Agriculture could be one of the main drivers.
In order to disseminate the results of this work, I spoke at 11 international conferences, workshops, and symposia during the two years of the fellowship. We expect to have four manuscripts resulting from this project, two of which are in preparation for submission. I also spoke at 9 outreach events for public audiences, and wrote 7 blog posts related to the project, which were shared on social media.
