Network simulations to study seed exchange for agrobiodiversity conservation

Abstract

Crop diversity is essential for sustainable development because diverse crops cope better with disease and climate change. A way to maintain crop diversity is to sustain seed exchange among farmers. Network simulations help in establishing which network properties promote crop diversity conservation. Here, we modelled the likelihood that an introduced crop variety will spread in a seed exchange network. The network model is based on published data on a directed network of barley seed flows in seven villages of Northern Ethiopia. Results show that the number of households that can be reached when introducing a new variety depends on the number of outgoing links of the household that first received the new variety. The distribution of the number of both incoming and outgoing links shows a departure from a normal distribution. This trend is explained by the presence of a minority of highly connected households and of a majority of weakly connected households. For the whole network, there is no significant correlation between the number of incoming and outgoing links of households. The findings explain the common observation that individual farmers do not cultivate all varieties present in a seed system. Absence of reciprocal exchange makes such networks less vulnerable to wholesale displacement of farmer varieties by improved ones.