Context Rusa deer (Cervus timorensis), originally introduced in the 1860s, are still spreading in eastern Australia. The expanding peri-urban rusa deer population in the Illawarra region of New South Wales, Australia is having undesirable impacts on human and ecological communities, but the spatial structure of this population has not been investigated. Genetic information on invasive species is potentially useful in identifying management units to mitigate undesirable impacts. Aims The aim of this study was to investigate population structure, characterise dispersal, and determine if natural and human-made landscape features affected gene flow in rusa deer invading the Illawarra region of New South Wales. Methods We used reduced representation sequencing (DArT-Seq) to analyse single nucleotide polymorphisms distributed throughout the genomic DNA of rusa deer culled during a management program. We used admixture and Principal Component Analyses to investigate population structure with respect to natural and human-made landscape features, and we investigated whether our genetic data supported the presence of sex-biased dispersal. Key results Genetic diversity was highest in the north, near the original introduction site. A railway line demarcated restricted gene flow. Surprisingly, the Illawarra escarpment, a prominent landscape feature, did not restrict gene flow. There was no evidence of sex-biased dispersal and seven individuals were identified as genetic outliers. Conclusions The genetic structure of the Illawarra rusa deer population is consistent with individuals spreading south from their introduction site in Royal National Park. The population is not panmictic, and a landscape feature associated with urbanisation was associated with increased spatial genetic structure. Outliers could indicate hybridisation or secondary incursion events. Implications Rusa deer can be expected to continue invading southwards in the Illawarra region, but landscape features associated with urbanisation might reduce dispersal across the landscape. The genetic structuring of the population identified three potential management units on which to prioritise ground shooting operations.