Model description

The emerald ash borer (EAB; Agrilus planipennis Fairmaire) is a highly destructive invasive pest of ash (Fraxinus spp.), responsible for the mortality of millions of trees in regions where it is non-native. Although EAB is not currently established in Great Britain (GB), its potential arrival poses a significant threat to native ash (Fraxinus excelsior L.), which is already under pressure from ash dieback (ADB; Hymenoscyphus fraxineus). Consequently, the development of effective surveillance and early-detection strategies for EAB is essential.

This repository contains a spatially explicit, stochastic model of EAB introduction, spread, and detection in Great Britain. The model integrates three key components:

(i) The estimated spatial prevalence of ash dieback,

(ii) The population dynamics and dispersal of EAB following arrival, and

(iii) A socio-dynamics module that simulates land-manager behaviour with respect to surveillance and tree management, based on a values-driven decision-making framework.

If EAB is detected within the model, a contingency response is triggered that includes felling of infested trees and intensified visual surveillance, with the potential to eradicate local outbreaks or slow pest spread. The model is used to evaluate the effectiveness of alternative surveillance strategies, including targeted trapping at high-risk sites, routine inspections by land managers, and volunteer-based surveillance, with or without subsidised trapping.


Model implementation

The model is currently configured to run on a 300 m × 300 m spatial grid for Suffolk. However, all required data are also provided to run simulations for Kent and Wales. To change the case-study region, users must update the grid dimensions (lines 282–283 in EABGrid.cpp) and modify the hard-coded input paths (lines 938–971 in EABGrid.cpp).

The model is written in C++ and was compiled using Microsoft Visual Studio 2022. Stochastic sampling and numerical integration of the dispersal kernel use routines from the NAG Library (https://nag.com/nag-library/). These dependencies can be replaced with alternative numerical libraries if a NAG licence is not available.

Simulations are run at annual time steps over a 15-year period.

Outputs

Model outputs are written to files in the Outfiles directory. This directory includes example MATLAB scripts to visualise simulation results. An Excel file is also provided containing the outputs used in Alonso Chavez et al. (2026).

Funding and acknowledgements

This code was developed as part of the NERC-funded SMARTIES project (NE/T007729/1). Rothamsted Research receives strategic funding from the Biotechnology and Biological Sciences Research Council (BBSRC), UK. Additional support was provided by the Growing Health Institute Strategic Programme (BB/X010953/1; BBS/E/RH/230003C).

References

Alonso Chavez, V., Brown, N., Van den Bosch, F., Parnell, S., Dyke, A., Hall, C., Karlsdottir, B., Marzano, M., Morris, J., O’Brien, L., Williams, D., & Milne, A. E. (2025). Early detection strategies for invading tree pests: Targeted surveillance and stakeholder perspectives. Journal of Applied Ecology, 1–15. https://doi.org/10.1111/1365-2664.70009

Alonso Chavez, V., Brown, N., Parnell, S., Coombes, M., Dyke, A., Hall, C., Karlsdottir, B., Marzano, M., Morris, J., O’Brien, L., Williams, D., 
Milne, A.E. (2026) Surveillance of Ash Trees Under Multiple Threats: Integrating Emerald Ash Borer and Ash Dieback Dynamics with Stakeholder Behaviour. Journal of Applied Ecology.