The choice of a modeling approach is a critical decision in the modeling process, as it determines the complexity of the model and the phenomena that the model captures. In this paper, we developed an individual-based model (IBM) and compared it to a previously published ordinary differential equation (ODE) model, both developed to describe the same biological system although with slightly different emphases given the underlying assumptions and processes of each modeling approach. We used both models to examine the effect of insect vector life history and behavior traits on the spread of a vector-borne plant virus, and determine how choice of approach affects the results and their biological interpretation. A non-random distribution of insect vectors across plant hosts emerged in the IBM version of the model and was not captured by the ODE. This distribution led simultaneously to a slower-growing vector population and a faster spread of the pathogen among hosts. The IBM model also enabled us to test the effect of potential control measures to slow down virus transmission. We found that removing virus-infected hosts was a more effective strategy for controlling infection than removing vector-infested hosts. Our findings highlight the need to carefully consider possible modeling approaches before constructing a model.
- Barley yellow dwarf virus
- Individual-based model
- Mean field
- Ordinary differential equation
- Vector-borne plant pathogen