Modelling RCD
Summary Final Report - Operational Research 1996/97
Business/Institution: AgResearch
Programme Leader: Dr Nigel Barlow
Programme title: Modelling RCD
Goal and context:
The goal was to increase understanding of the epidemiology and impact of RCD in New Zealand using models. Lack of knowledge about the disease was a principal reason for the decision not to release it in New Zealand, and still remains an important limitation on our ability to manage it and to determine whether we can manage it. The work built on previous MAF-funded model development and allowed continuity between this and possible forthcoming PGSF-funded work.
Brief outline of methodology:
The approach used non-spatial, 'classical' or Anderson/Maytype disease/host models, with the rabbit population dynamics treated in differing levels of detail. A novel model variant was developed to account for apparent observed RCD behaviour which could not be explained by the conventional models.
Results and Discussion:
A model of intermediate complexity was completed and accepted for international publication. This model provided new understanding of the factors governing the character, speed and recurrence of RCD epidemics, and suggested management conclusions for scenarios in which the disease either persisted or did not. It suggested that the rapid, intense epidemics observed in Australia were inconsistent with local disease persistence, and therefore that the latter was unlikely. Since then, two mechanisms have been found which reconcile rapid epidemics with persistence, suggesting that 'classical' biological control from RCD is a distinct possibility.
One of these mechanisms involved a novel model with multiple disease transmission pathways, another involved revised serology data. With RCD persistence, the models suggested:
- yearly epidemics in summer or autumn,
- sustained suppression of high-density Otago rabbit populations of 75 - 95%,
- low prevalences of the disease (< 5% infected) during repeat epidemics,
- little effect of seasonality from vector activity (the seasonal pattern of RCD appears to be determined at least as much by the intrinsic dynamics of the rabbit/RCD system as by seasonality in vector activity),
- that care is needed in applying additional controls because these can eliminate otherwise persistent disease if rabbit density is reduced too much,
- persistent disease in low density lowland populations as well as in high-density Otago ones, given certain disease scenarios, but with reduced levels of suppression,
- that rabbit suppression from initial epidemics may be density-dependent, though the initial knockdown due to direct infection from baits will be density-independent (as for any other control), and 8) that spotlight counts should be undertaken at least monthly if the data are to contribute usefully to understanding RCD epidemiology.
Further work in modelling should centre on:
- analysis of a completed and more detailed RCD/rabbit population model to test robustness of conclusions from the intermediate complexity model,
- parameterisation of the models from New Zealand field data, especially assessment of local persistence of disease and which mechanism of the two identified is responsible,
- incorporation of a more detailed and explicit treatment of vector effects based on field data,
- use of the models to optimise RCD management, and
- development of spatial models for RCD spread and non-local persistence.
Contact for Enquiries
Manager, Strategic Science Team
MAF Biosecurity New Zealand
PO Box 2526
Wellington
NEW ZEALAND
Phone: +64 4 894 0115
Fax: +64 4 894 0731
Contact this person
