Skip to main content
ARS Home » Southeast Area » Fort Pierce, Florida » U.S. Horticultural Research Laboratory » Subtropical Plant Pathology Research » Research » Publications at this Location » Publication #304307
Title: Optimising and Communicating Options for the Control of Invasive Plant Disease When There Is Epidemiological Uncertainty

Author
item CUNNIFFEL, N - University Of Cambridge
item STUTT, O - University Of Cambridge
item DESIMONEL, R - University Of Cambridge
item Gottwald, Timothy
item GILLIGAN, C - University Of Cambridge

Submitted to: PLoS Computational Biology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/25/2015
Publication Date: 4/13/2015
Citation: Cunniffel, N.J., Stutt, O.J., Desimonel, R.E., Gottwald, T.R., Gilligan, C.A. 2015. Optimising and Communicating Options for the Control of Invasive Plant Disease When There Is Epidemiological Uncertainty. PLoS Computational Biology. 11(4)e1004211. https://doi.org/10.1371/journal.pcbi.1004211.
DOI: https://doi.org/10.1371/journal.pcbi.1004211

Interpretive Summary: We have developed and interactive, web-based model that analyses the factors underlying the effectiveness of eradication programs/strategies of small-scale outbreaks of invading plant diseases. To do this we have used the eradication program for citrus canker in Florida (1996-2006) as a case study. However, the model is quite adaptable and can be modified to address any plant pathogen or insect pest for which adequate data exist. The model demonstrates how to optimise control by removal hosts surrounding detected infection. We can also use the model to optimize control to require a minimum of time or removal of a minimum of plant hosts. The model is available online as an interactive user-friendly interface at http://www.webidemics.com/. The model is designed to be used by regulatory officials and stakeholders as well as researchers to explore the optimization of eradication while minimizing socio-political impact.

Technical Abstract: We analyse factors underlying the effectiveness of reactive eradication of small-scale outbreaks of invading plant disease. Our results are generic, although we use the eradication program for citrus canker in Florida (1996-2006) as a motivating case study. We use a spatially-explicit stochastic epidemiological model to demonstrate how to optimise control via removal of hosts surrounding detected infection (i.e. localised culling). Our model is available online as an interactive user-friendly interface at http://www.webidemics.com/. We show how optimal culling strategies can be defined, even though stochasticity in disease spread leads to a distribution of epidemic impacts for fixed control and disease spread parameters. We show how performance of control depends on epidemiological parameters controlling pathogen spread, the initial level of infection, and the logistics and implementation of detection and control. We also consider how controls are conditioned on the level of risk deemed acceptable, and show how to account for potential global impacts of a small-scale outbreak. Eradication by culling can be effective, particularly when it is started quickly. However the optimum strategy and its performance are both sensitive to parameters controlling pathogen spread and control implementation, together with the level of risk aversion.