Modelling the 1996 lupin anthracnose incursion in Western Australia

In this presentation I will explain how we used computational modelling to investigate the spread of lupin anthracnose over a spatially heterogeneous landscape. I will consider a situation analogous to the 1996 incursion in Western Australia which crippled the albus lupin industry, and address the question of whether the incursion could potentially have been eradicated. I will focus on several important factors that influence the likelihood of eradication: spatial characteristics of landscape suitability, organism detectability, surveillance effort and the type of management strategy applied.

For the purpose of this case study, a spatially heterogeneous suitability map that describes this landscape was created using a satellite image. I will discuss how the method used for creation of the suitability map has applications in a rapid response protocol and is part of the general system to rapidly classify potential spread of any new disease or insect invader that is being developed in CRCNPB project 10124. I will also demonstrate the effect that the spatial pattern of lupin paddocks has on disease progression and detection.

Within the simulation, we use a simple detection rule where the probability of detection is a function of the detectability of the organism and the search effort. Upon detection, control techniques are implemented. The predicted success of different control options and how this depends on organism detectability and surveillance effort under different scenarios will be discussed.