The directed network resulting from the between-herd cattle movements in France will be described using graph theory to identify territorial specificities and temporal structures potentially occuring in the network. Moreover, stratified analysis will be performed based on either characteristics (such as age and race) of animals who are moved between farms or of the types of origin and destination herd (dairy, beef) to explore the existence of different network topologies according to these features. Data ranging from 2005 to 2009, available at the animal level from a national database of cattle identification (BDNI), will be analysed.
This network will be integrated in epidemiological models of Cb, Map and BVDV spread. Regarding Cb, the model will account both for transmission through animal movements and for airborne dispersion, depending on results of Task 2a. The relative impacts of meteorological conditions, between-herd trade intensity, and between-herd distance will be assessed under different levels of infection prevalence and in farming systems of Western France and Sweden. Field data collected in a follow-up study as well as data drawn from the literature will be used to calibrate the model. The model will be used to assess the zoonotic risk in urban areas neighbouring infected dairy herds. Regarding Map and BVDV, as transmission mainly occurs through direct contacts between animals, the only between-herd transmission route considered will be animal movements. A model of between-herd BVDV spread has been developed for a small metapopulation (Courcoul and Ezanno, 2010). It will be improved to represent a region and to account for different types of herds (dairy vs. beef, with vs. without fattening activity) because several types of herds interact in a region and within-herd spread varies with herd structure (Ezanno et al., 2008). Regarding Map, the between-herd model will be based on a within-herd model already developed (Marcé et al., in revision a, b). We will assess the influence of network characteristics on pathogen spread. These models then will be used to evaluate control strategies at a regional scale (WP3), focusing on the control of the health status of source herds for animal purchases.