Combined application of social network and cluster detection analyses for temporal-spatial characterization of animal movements in Salamanca, Spain

Abstract

Social network analysis was used in combination with techniques for detection of temporal-spatial clusters to identify operations at high risk of receiving or dispatching pigs, from January through December 2005, in the Spanish province of Salamanca. The temporal-spatial structure of the network was explicitly analyzed to estimate the statistical significance of observed clusters. Significant (P<0.01) temporal-spatial clusters identified were grouped into two compartments based on the nature and extent of the contacts among operations within the clusters. One of the compartments was identified from January through April, included a high proportion of extensive farms (0.39), and was likely to be related with the production and trade of Iberian pigs. The other compartment encompassed a smaller proportion of extensive farms (0.11: P<0.01), took place from May through December, and was probably related to intensive production systems. Analysis of a sub-section of the network, which was selected based on the administrative division of Spain, yielded to the identification of a different set of clusters, showing that results of social network analysis may be sensitive to the extension of the information used in the analysis. The approach presented here will be useful for the implementation of differential surveillance, prevention, and control strategies at specific times and locations, which will aid in the optimization of human and financial resources.