An agent-based simulation modelling approach to assessing various refuge strategies in Bt sugarcane

Journal Title
Journal ISSN
Volume Title
Stellenbosch : Stellenbosch University
ENGLISH SUMMARY : Crops expressing genes from the bacterium Bacillus Thuringiensis (Bt) produce a protein toxic to members of the order Lepidoptera and are a popular alternative to sprayed insecticides. Although these Bt crops are considered to be an effective pest control method, careless usage may add selective pressure on the pest population to develop resistance to the protein over time. One method of limiting the rate of resistance development is to keep small portions of the cultivated land planted with the non-GMO crop which then acts as a refuge area for susceptible pests, limiting its exposure to the protein. Varieties of Bt sugarcane for the South African market that should limit the damage caused by the stalk borer moth, Eldana Saccharina Walker, are being developed, and a prerequisite to releasing such a product is to manage resistance development. Refuge areas are the primary way of managing resistance, but a recommendation is required on the size and layout of the refuge areas to be planted, as an area too small may not curb the rate of resistance development enough, but an area too large may not be economically viable for the industry. An agent-based simulation model is presented where individual moths are modelled as agents on an underlying sugarcane field that can either be Bt or refuge. The spatial aspect of the simulation model allows for testing the effectiveness of different shapes of refuge and not just the size. To reduce the complexity of the model, the field is divided into a series of smaller cells that can interact with each other, which allows us to model the impact of a severe infestation on an area the size of an average sugarcane farm. The simulation divides the simulated space into two grids representing the layers of the sugarcane that the different life stages of the insect move in. The first layer is the ground layer, representing the lower portions of the sugarcane where the eggs, larvae, pupae and egg-laying female moths are found. These cells are smaller as the immature life stages are nearly stationary and the female moth is assumed to have a much lower dispersal rate when searching for sights to lay eggs. The second layer is the sky layer with larger cells, where adult males and unmated females are able to move over larger distances to find mates. To further reduce complexity, every cluster of eggs laid by a female moth is modelled as a single agent that progresses as a unit through the immature life stages before being converted to many individual moth agents when they mature. The model is then applied to three hypothetical case studies, each focussing on a specific aspect of refuge planning. The first case study focusses on the size and distribution of refuge, the second on the shape of the refuge, and the third assumes that the Bt sugarcane is introduced gradually and how this may affect the refuge strategy. The results from these case studies as well as those obtained from the sensitivity analysis are then summarised in the form of a decision support tool (DST), which provides some recommendations for refugia planning and highlights the risk factors for resistance development identified during the course of this study. Combining the recommendations with the risk factors that may be associated with a given scenario should provide the growers with a starting point of how to plan their refugia.
AFRIKAANSE OPSOMMING : Raadpleeg teks vir opsomming.
Thesis (MCom)--Stellenbosch University, 2019.
Refuge Areas -- Simulation methods, Sugarcane -- Diseases and pests, Sugarcane borer -- Resistance to toxin, UCTD