Social-ecological modelling of potato late blight : Managing crop resistance in disease control

Abstract

Social-ecological systems are a case of complex adaptive systems. They consist of many components that interact across temporal and spatial scales. The system’s overall behaviour emerges from the interactions between the different components, while these components adapt to each other and to the environment. In this thesis, potato late blight control is used as a model system. Systems methods are used to analyse its behaviour and contribute to the development of sustainable disease management strategies. Late blight, caused by Phytophthora infestans, is one of the main threats in potato production worldwide; in the Netherlands, high potato density and favourable weather conditions result in frequent outbreaks of the disease. Crop resistance could play an important role in sustainable disease control by reducing chemical fungicide inputs. However, due to the rapid adaptation of P. infestans, there is a risk that the resistance is overcome by the pathogen. The overall disease incidence in the landscape is influenced by spatial biophysical processes as well as crop management strategies of farmers who are influenced by their socio-institutional environment. To analyse the social-ecological interactions a collection of complementary methods was used. In-depth interviews with farmers, breeders and other experts increased understanding of the Dutch potato sector and of factors involved in decision making on late blight control. Fuzzy cognitive mapping showed that social and ecological processes are tightly linked by various feedback loops. Both an increase in stakeholder cooperation and a change in market demands towards resistant cultivars could improve sustainability of late blight management. In contrast, policies restricting the use of fungicides would result in increased disease severity if no alternative strategies were implemented, which would require social-institutional support and facilitation. Agent-based modelling (ABM) was used to (i) analyse crop-disease interactions affected by management strategies at the landscape level, (ii) analyse social-ecological interactions between farmer behaviour and disease dynamics and (iii) communicate the dynamics in the complex system to stakeholders. Results showed that increasing the fraction of resistant potato fields strongly reduced late blight incidence in a landscape. However, resistance breakdown could occur by emergence and spread of a new virulent strain. It was found that low as well as high proportions of fields with the resistant variety could increase durability of resistance. The ABM showed that resistance breakdown is the result of interactions between management strategies of farmers, the weather conditions and the allocation of potato varieties in the landscape. Several resistance management strategies were identified that could potentially be effective to increase resistance durability, including (reduced) use of fungicides on all resistant or all susceptible fields, growing a resistant variety with multiple resistance genes (instead of single-gene resistant varieties) and immediate haulm destruction of resistant fields after infection with the virulent strain. ABM-based scenarios were used as learning tool in workshops with farmers. Scenarios of disease dynamics at the landscape level increased awareness that collective action is needed to prevent emergence and spread of virulent strains. Significant differences in perceptions on disease control were found before and after the workshop as well as between organic and conventional farmers. It was concluded that the workshop increased farmers’ knowledge of the system and served as a good starting point for discussions among actors.