Assessing the availability of pollinator foraging resources using remote sensing technology

Barnsley, Sarah (2021) Assessing the availability of pollinator foraging resources using remote sensing technology. Doctoral thesis, University of East Anglia.

[thumbnail of SB 230206 Final PhD Thesis.pdf] PDF
Restricted to Repository staff only until 31 March 2026.

Request a copy


Nectar sugar is the primary energy source for pollinators and an integral part of their diet. The supply of nectar sugar available to pollinators has been quantified at habitat and landscape scales, but few studies have explored how nectar sugar varies within habitat types, at finer scales. This thesis quantifies fine-scale variability in nectar supply in linear habitat features managed for pollinator conservation at a UK arable farm, and explores for the first time whether this variability can accurately be measured using remote sensing.

In a study of sown and naturally regenerated field margins, I find substantial variation between margin types and individual margins in the quantity of nectar sugar supplied (mg/m2/day), and a significant positive relationship between nectar sugar supply and honeybee abundance.

Baseline assessment of the spatial and temporal distribution of nectar sugar across a farmed landscape is necessary to identify gaps in supply and inform pollinator conservation strategies. Such assessment must include variability within habitat types demonstrated in this thesis. Traditional ground-based quadrat surveys are time consuming, resource-intensive and require expertise. An alternative is to use remotely-sensed imagery to classify nectar-rich flowering plant species. I demonstrate high overall accuracies (92.3%-98.7%) and variable user’s and producer’s accuracies for 11 nectar-rich flowering plant species, using multispectral imagery (spatial resolution 3-7cm) obtained via manned aircraft and unmanned aerial vehicle platforms. I also demonstrate that two of the major nectar producers, Centaurea nigra and Cirsium arvense, are more accurately classified when combined as a single category (Thistgroup) and that area classified as Thistgroup is correlated with the mean number of floral units on the ground (units/m2).

The methods developed in this thesis have multiple applications for pollinator conservation in agricultural settings, providing they are scaleable. The final Chapter discusses these possible applications and next steps for research.

Item Type: Thesis (Doctoral)
Faculty \ School: Faculty of Science > School of Biological Sciences
Depositing User: Chris White
Date Deposited: 20 Mar 2023 15:11
Last Modified: 20 Mar 2023 15:11

Actions (login required)

View Item View Item