Alternative Definitions of Floral Resource Availability Alter Inferred Plant Importance in Plant-Pollinator Networks

Area of Science:

• Ecology
• Entomology
• Botany

Background:

• Null models are vital for inferring pollinator preferences from plant-pollinator networks.
• Inferences are sensitive to how floral resource availability is defined, especially for plants with complex structures.

Purpose of the Study:

• To investigate how different definitions of floral resource availability impact null model inferences in plant-pollinator networks.
• To test alternative definitions of floral resource availability using *Heracleum sphondylium* (common hogweed).

Main Methods:

• Collected plant-pollinator interaction data from three field sites near Bristol, UK.
• Quantified common hogweed floral abundance using three definitions: individual flowers, floral units, and umbels.
• Applied null models to analyze the visitation data with each resource definition.

Main Results:

• The inferred importance of common hogweed varied significantly based on the resource definition used.
• Hogweed was often classified as a keystone plant when using floral units or umbels, but not consistently with individual flower counts.
• Altered resource definitions changed the inferred importance of other plant species and broader pollinator preference patterns.

Conclusions:

• Floral resource definitions critically influence ecological inferences drawn from plant-pollinator visitation data.
• Explicitly defining and reporting floral resource units is essential for accurate ecological interpretation.
• This highlights the need for standardized methods in plant-pollinator network research.