Modelling of brown rot prevalence in the Dutch potato production chain over time: from state variable to individual-based models

The state-variable model proved useful for obtaining insight into the basic principles of brown rot dispersal. It showed that the dynamics of the fraction of infected seed lots in the total potato lot population forms the key to a general understanding of brown rot epidemics. However, this model was unable to reflect the large fluctuation in yearly number of infections that is inherent to brown rot epidemics. To give a more detailed and realistic representation of the fraction of infected seed lots, a conceptual IBM was developed. As in this IBM a specific location is assigned to each individual potato lot, it becomes straightforward to include spatial heterogeneities based on detailed data on the potato production sector. In contrast to the state-variable model, the IBM enables us to study the effects of specific brown rot control policies in spatially defined areas. Moreover, the inherent high level of detail makes the IBM a convenient technique for policy application. The IBM will be further developed and extended to a bio-economic model for application in brown rot control strategy analysis.