Identifying individual subsurface faults in a larger fault system is important to characterize and understand the relationship between microseismicity and subsurface processes. This information can potentially help drive reservoir management and mitigate the risks of natural or induced seismicity. We have evaluated a method of statistically clustering power spectra from microseismic events associated with an enhanced oil recovery operation in southeast Utah. Specifically, we were able to provide a clear distinction within a set of events originally designated in the time domain as a single cluster and to identify evidence of en echelon faulting. Subtle time-domain differences between events were accentuated in the frequency domain. Power spectra based on the Fourier transform of the time-domain autocorrelation function were used, as this formulation results in statistically independent intensities and is supported by a full body of statistical theory upon which decision frameworks can be developed.