Developing models for the spread and management of National Priority Plant Pests
This project will produce mechanistic and statistical models to support the management of NPPP incursions. Eradication and containment models will be based on plausible pest establishment and detection scenarios in operational settings. Managing incursions requires that knowledge of pest ecology/epidemiology will work in conjunction with surveillance data to guide the appropriate zoning and implementation of control measures. Models will simulate the spread of incursions from potential establishment locations through natural and human-assisted spread. The capacity for surveillance data to delimit incursions with respect to control technologies will be determined through statistical modelling.
The Department is the major contributor of resources to eradication and containment activities and plays a coordinating role in early detection surveillance for National Priority Plant Pests (NPPP). Effective deployment of resources for early detection surveillance will pre-emptively lower Australia’s potential liability for incursion costs.
Emergency responses to major pests consume significant resources which can be reduced by a more informed understanding of the relationship between pests, the incursion environment and surveillance information. Modelling can provide guidance to the CCEPP, NMG and advisory groups on the appropriate course of action for response management, including technical feasibility and the cost benefit of eradication or containment. Managing incursions of priority plant pests is often confounded by a poor understanding of the distribution of the pest. Surveillance activity tries to refine the potential distribution over time, but it can be difficult to understand the hidden incursion process in relation to presence and absence data, particularly for pests with broad host ranges, complex spread pathways and poor detectability. Custom-made models have been constructed in response to emergency plant pest incursions in the past, but the Australian Animal Disease (AADIS) model (Bradhurst et al., 2015), will provide the basis for a better maintained departmental system that will help prepare for high priority pests, as well as being adaptable for use in responses to other pests.
This project will produce mechanistic and statistical models to support the management of NPPP incursions. Eradication and containment models will be based on plausible pest establishment and detection scenarios in operational settings. Managing incursions requires that knowledge of pest ecology/epidemiology will work in conjunction with surveillance data to guide the appropriate zoning and implementation of control measures. Models will simulate the spread of incursions from potential establishment locations through natural and human-assisted spread. The capacity for surveillance data to delimit incursions with respect to control technologies will be determined through statistical modelling.
Year 1 saw significant changes to the AADIS software architecture. The model now operates in three distinct modes: livestock disease, vector-borne disease and plant pest. As suggested during the project workshop held in Canberra on 29th August 2017, an exotic fruit fly (Bactrocera dorsalis) was chosen for the initial case study.
During year 2 of this project control measures (detection, surveillance, movement restrictions, destruction and treatment) will be implemented as per the PLANTPLAN (Plant Health Australia, 2016). The project will investigate the spatial representation of the host material over time in relation to incursions, including the availability of commercial, backyard and weed hosts. Model stability over different scales will be assessed and methods to address parameterisation will be documented.
The model will proceed with fruit fly trapping as a case study to identify control and trapping scenarios and their sensitivity to assumptions about dispersal and lure attractancy. Further development of habitat layers will be pursued in order to develop more realistic spread and control dynamics. Model sensitivities to assumptions surrounding the declaration of eradication will be explored.
