# Modelling the initial stages of biocontrol of the invasive herb Tradescantia fluminensis by beetles

## TL;DR

This paper models how beetles can control the invasive plant Tradescantia fluminensis, finding that large initial beetle populations may be needed for effective control.

## Contribution

A novel Monte Carlo method is used to estimate beetle release requirements and assess sensitivity in biocontrol modeling.

## Key findings

- Large initial beetle populations may be needed for effective plant control.
- Biocontrol success depends on specific beetle-plant interaction parameters.
- The model can be expanded to evaluate long-term interactions.

## Abstract

Tradescantia fluminensis is a herbaceous monocot originally from South America which is an invasive weed in many countries. Manual control is highly intensive and efforts have been made to establish colonies of three biocontrol beetles: Neolema ogloblini, Neolema abbreviata and Lema basicostata. Each beetle has its own characteristic life cycle and pattern of tradescantia damage. We construct a model of both plant and beetle population to assess the initial stages of interaction after the release of each biocontrol species. We parametrize the model using field data and give values for parameters which are hard to observe. We use the model to estimate the size of the released population needed for immediate plant control and test the sensitivity of this estimate to parameter estimates using a novel Monte Carlo method. Our results show that this may be a successful method of biocontrol under some circumstances, but the initial number of beetles released would be unrealistically large in some cases. We consider ways that this model could be expanded in future to evaluate longer-term interactions.