Mapping nectar-rich pollinator floral resources using airborne multispectral imagery.

Area of Science:

• Ecology
• Remote Sensing
• Conservation Biology

Background:

• Wild pollinator populations are linked to flower-rich habitats.
• Agricultural systems often lack sufficient floral resources for pollinators.
• Understanding resource availability is key for effective habitat management.

Purpose of the Study:

• To assess the potential of very high-resolution remote sensing for mapping pollinator foraging resources.
• To identify temporal and spatial gaps in floral resources within agricultural landscapes.
• To develop a remote sensing approach for monitoring pollinator conservation.

Main Methods:

• Used multispectral airborne imagery (3cm and 7cm resolution) in a UK agricultural landscape.
• Classified five key nectar-rich flowering plant species using a maximum likelihood algorithm.
• Acquired imagery in March, May, and July to capture seasonal variations.

Main Results:

• Achieved overall classification accuracies above 90% for all months and resolutions.
• Demonstrated that higher spatial resolution does not always guarantee higher accuracy.
• Identified a prototype approach for mapping pollinator resources in agricultural settings.

Conclusions:

• Very high-resolution remote sensing is a viable tool for mapping pollinator foraging resources.
• The developed method provides a foundation for a remote sensing pipeline to monitor nectar-rich plant availability.
• Further research is needed to distinguish co-flowering species and quantify floral density for nectar supply calculations.