12–17 Jul 2026
University of Graz
Europe/Vienna timezone

Lightning Laplace methods for floral-arthropod electrostatics

16 Jul 2026, 14:40
20m
15.34 - SR (University of Graz)

15.34 - SR

University of Graz

40
Contributed Talk Multiscale and Multiphysics Modelling Contributed Talks

Speaker

Samuel Harris (University of Bristol)

Description

Insects have acquired a variety of sensory receptors through coevolution with plants, which transmit floral cues such as scent, colour and shape. Recently discovered among this suite of arthropod senses is electroreception – the ability to detect electric fields. This raises the question of whether plants also convey biologically relevant information electrically to nearby pollinators, and how they interact electrically with each other and the wider environment. To investigate this, the electric field is modelled numerically using a two-dimensional approximation. A variant of recently developed lightning Laplace methods known as the AAA-least squares algorithm computes a rational approximation of the harmonic electric potential with high accuracy and speed, even on a standard laptop. This numerical approach is extended in two non-trivial ways: first to two-domain problems to solve the electric field inside and outside a single flower; and second to general multi-domain problems representing an entire flower meadow. These new extensions are presented in detail, followed by some biologically motivated analysis of floral morphology, pollen load and various plant-arthropod configurations. Comparison with results produced in COMSOL shows that the two-dimensional results give a good proxy of the overall three-dimensional behaviour.

Bibliography

article{harris2025modeling,
title={Modeling floral and arthropod electrostatics using a two-domain {AAA}-least squares algorithm},
author={Harris, S.J. and Palmer, R.A. and McDonald, N.R.},
journal={SIAM J. Appl. Math.},
volume={85},
number={2},
pages={916--944},
year={2025},
publisher={SIAM}
}

Author

Samuel Harris (University of Bristol)

Co-authors

Robb McDonald (University College London) Ryan Palmer (University of Bristol, UK)

Presentation materials

There are no materials yet.