摘要Background Hemorrhagic fever with renal syndrome (HFRS) is endemic in Junan county, Shandong Province, China.We conducted geographic information system (GlS)-based spatial analysis with the objective of estimating the spatial distribution of rodent populations and their hantavirus infection patterns, to describe the spatial relationships of hantavirus strains in small ecological areas and to identify key areas in endemic areas of HFRS for future public health planning and resource allocation.Methods Rodent sampling was conducted in seven villages in Junan county from February 2006 to January 2007 using field epidemiological surveillance. Dynamics of hantavirus infection and population densities in rodents were investigated. Spatial statistical techniques including Ripley' L index and nearest neighbour hierarchical (NNH) clustering analysis were conducted to reveal the spatial structure of rodent populations in seven villages. Phylogenetic analysis and two-dimensional minimal spanning tree (2-D MST) models were employed to describe the spatial relationship of hantavirus strains.Results Data showed that Mus musculus was the most common species in our study area, followed by Rattus norvegicus. Ripley's L index and NNH analysis showed that the spatial distribution of all captured rodents, Mus musculus and Rattus norvegicus in seven villages were clustered and there were hotspot areas of rodent distribution. The branches of 2-D MSTs had similar topologies to those of corresponding phylogenetic trees, and hantavirus strains exhibited obvious connective traces in seven villages.Conclusions These results contribute to the understanding of the spatial distribution of rodent populations and hantavirus infection patterns in small areas, and identify priority areas within the epidemic areas for the development of a better prevention strategy against hemorrhagic fever with renal syndrome in a small ecological area.