In order to study microbial diversity in a polycyclic aromatic hydrocarbon-impacted soil, 14 bacterial strains were analyzed by 16S rRNA gene sequencing and amplified fragment length polymorphism (AFLP) analysis. Bacterial strains isolated from two different hydrocarbon-polluted sites were identified to the species level by 16S rRNA full-gene sequencing using MicroSeq 16S rRNA gene sequencing. Their genome was subsequently analyzed by high-resolution genotyping with AFLP analysis, in order to monitor species variability and to differentiate closely related strains. Cluster analysis based on AFLP fingerprinting showed intra-specific polymorphism, even among strains with 100% 16S rRNA gene sequence identity. The results show that AFLP is a powerful, highly reproducible and discriminatory tool for revealing genetic relationships in bacterial populations. The ability to differentiate and track related closely microbes is fundamental for studying structure and dynamics of microbial communities in contaminated ecosystems.