Female visitors construct the robust architectural core of flower-visitor networks

Ecological community stability is a crucial topic in ecology, often assessed through coextinction models. However, these models can be influenced by artificial factors like network size and connectance. Furthermore, the impact of visitor sex on network structure is often overlooked. This research examined the actual structural resilience of flower-visitor networks by dividing them into female and male subnetworks.

To eliminate size-related biases, the study standardized robustness metrics against both simulated visitor and flower losses using two separate null models. While initial measurements incorrectly indicated full networks were the most resilient, adjusting these measurements based on null models uncovered a significant discrepancy.

When faced with visitor extinctions, female subnetworks exhibited substantially greater structural robustness than the entire networks. This female-led resilience remained even after the exclusion of highly numerous eusocial bees. However, this resilience was absent when considering flower losses, highlighting a trophic imbalance where plants serve as critical hubs.

This study provides the first concrete evidence that female foraging behaviors form the resilient structural backbone of pollination communities. The researchers extend their gratitude to Dr. Taro Maeda for insightful discussions and feedback on an earlier version of the paper. The supercomputer "Shiho" at the National Agriculture and Food Research Organization was utilized for null model calculations.

This research was funded by JSPS KAKENHI grants JP23K05281, JP26K09071, and JP26K09087. The research was conducted at the Research Center for Agricultural Information Technology, National Agriculture and Food Research Organization in Tsukuba, Japan. The article is licensed under a Creative Commons Attribution 4.0 International License, allowing for use, sharing, adaptation, distribution, and reproduction in any format, provided proper credit is given to the original authors and source, and a link to the Creative Commons license is included.