Estimation of national-level farm demographic data for preparedness of highly infectious livestock disease epidemics
Project ID: 1502C (1402C)
CEBRA Project Leader: Mark Burgman
DA Sponsor: N/A
DA Project Leader: NZ MPI: Daan Vink, Mary van Andel
DA Division: NZ MPI
MPI Project Manager: Christine Reed
Collaborators: Tim Carpenter, Carolyn Gates, Masako Wada, Massey University EpiCentre; Tracey Hollings, CEBRA; Tom Kompas, Crawford Business School, ANU; Andrew Robinson, Dept of Mathematics and Statistics, University of Melbourne.
Epidemic spread models depend sensitively on initial conditions, including the distribution of livestock among farms. Project 1402C developed a suite of statistical models that estimate the number of animals on farms in New Zealand from remotely sensed data at a scale that is relevant to support emergency response planning. These models have identified areas where data are sparse and where uncertainties are relatively high. The economic impacts of the inaccuracies have not been modelled, and there is no way for assessing priorities for reducing these uncertainties.
Project 1502C will fill this knowledge gap. We will employ sensitivity analysis to answer this question in the New Zealand context and will use the results for economic modelling. We will consider four main categories of inaccuracies and a range of scenarios, including spatial inaccuracy, missing data, incomplete form information and uncertainty in animal number estimates. This project will quantify the impact of demographic and spatial data inaccuracy by using both epidemic and economic modelling, both at the farm and the macroeconomic level. The results will be used to set priorities to acquire additional data and to improve model fit, so that the economic costs of disease incursion can be minimized efficiently.