The fossorial water vole Arvicola terrestris is a grassland rodent and significant agricultural pest in the eastern French region of Franche-Comté. This species exhibits population outbreaks with a mean period of 6 years and a high abundance phase of 1-3 years. Little is known about the bacterial parasites found in this species or how bacterial parasite communities may change during the peak phase. We used high-throughput sequencing to sequence the V4 region of the 16s rRNA gene of bacterial DNA extracted from spleens, livers, lungs, hearts, and kidneys of A. terrestris individuals collected during autumn 2014, spring and autumn 2015 and spring 2016 from 5 sites in Franche-Comté to determine if bacterial parasite communities in this host vary spatially and temporally during the host high abundance phase. 32 bacterial operational taxonomic units (OTUs) corresponding to 12 parasitic bacterial genera were detected, including Bartonella sp., Mycoplasma sp., Leptospira sp., Mycobacterium sp., and Bordetella sp. Generalized linear modelling was used to identify significant predictors of OTU richness at the host individual (infracommunity) and host population (component community) scale; infracommunity richness ranged from 0 to 9 OTUs/animal and was best predicted by sampling site and date, with spring richness significantly lower than autumn richness. Bootstrapped component community richness ranged from 6.9 to 17.7 OTUs/population and was best predicted by local host abundance, with high-abundance populations hosting richer communities than low-abundance populations. Communities from autumn 2014 were significantly richer than spring communities despite low local host abundance at one site. Mantel tests using Jaccard and geographic distance indicate that infracommunity similarity, but not component community similarity, decay (weakly) with geographic distance. Characterization of infra- and component community richness and similarity at high abundance provides us with a foundation from which changes in bacterial parasite communities during the decline and low-abundance phase can be explored.