The use of participatory methods & simulation tools to understand the complexity of rural food security

Abstract

The realisation of food security is hindered by its inherently complex nature. Within less developed countries, a number of contributing factors act to undermine food security. At the level of rural households, livelihoods remain largely dependent upon agriculture. Structural factors such as small plot sizes, widespread soil degradation and a continued dependence upon rain fed agriculture, leave smallholders vulnerable to climatic shocks. A key objective of the project was to explore the potential role of simulation tools in understanding the complexity of sustainable food provision. The use of simulation techniques can act as a tool to explore the complex social, ecological and political factors affecting food availability, access and utilisation. Agent-based modelling (ABM) is one possible technique, which comprises a computerised simulation of agents located within an environment. Each agent provides a computational representation of a real-world actor, such as an individual, household or institution. Behaviour at the system level is an emergent property of the collective behaviour at the local level, resulting from the interactions between agents and the environment through predisposed rules. Depending on the type of rules they abide by, ABM can be categorised as abstract, experimental, historical or empirical. The aim here was to determine whether results from participatory exercises with smallholder farmers could be utilised in the construction of an empirical ABM. By taking a systems approach, ABM can highlight knowledge gaps, where data and understanding of feedbacks may be lacking. However, a shortage of effective protocols and architectures to represent agents and their interactions may limit their potential. Behavioural decisions and coping strategies of farmers in the face of drought formed the focus of this study. A participatory rural appraisal (PRA) exercise was designed to elicit greater understanding of smallholder responses to drought; as well as the perceived impact of government interventions in the form of input subsidies. PRA is a broad term that encompasses a wide range of data collection techniques that aim to involve local stakeholders in analysing their own situation. Tools include matrix scoring, seasonal calendars and wellbeing ranking exercises. Results from participatory fieldwork were successfully incorporated into an agent-based model of Malawian smallholders. However, scarce availability of data was identified as a key factor undermining model robustness. Initial implementation of the model found inferences could be made concerning the impact of policy upon household decision-making and food security. Overall the study provides fertile ground for future work. It is hoped that by integrating PRA exercises and ABM it will be possible to create a collaborative framework that promotes interaction between scientists, policy makers and stakeholders, alike

More information

Identifiers

ORCID for James G. Dyke: orcid.org/0000-0002-6779-1682

Catalogue record

Export record