This paper describes a new approach for spatial relative risk mapping as a tool for geographical risk assessment. The spatial epidemiological analysis is based on geographically referenced data about pseudorabies (Aujeszky's disease) virus infections at farm level in a region of high animal density in Germany at the beginning of the national eradication project. On the basis of serological findings 186 farms were classified as positive out of a total of 482 investigated farms listed in veterinary administrative registers. Geographical cluster analysis was used to identify two areas of high risk (RR = 2.4 and 3.3). Non-parametric density estimation was used to estimate the proportion of infected farms per square kilometre. Furthermore, the spatial relative risk function was approximated through the prevalence ratio defined by the ratio of the local prevalence and the overall prevalence of the farms outside the cluster regions. The corresponding approximated relative risk map indicates and quantifies a clear spatial pattern of disease frequency.