Biotechnology offers new ways to use biological processes as environmental sensors. For example, in soil microbial fuel cells (MFCs), soil electro-genic microorganisms are recruited to electrodes embedded in soil and produce electricity (measured by voltage) through the breakdown of substrate. Because the voltage produced by the electrogenic microbes is a function of their environment, we hypothesize that the voltage may change in a characteristic manner given environmental disturbances, such as the contamination by exogenous material, in a way that can be modelled and serve as a diagnostic. In this study, we aimed to statistically analyze voltage from soil MFCs injected with urea as a proxy for gross contamination. Specifically, we used 𝘬-means clustering to discern between voltage output before and after the injection of urea. Our results showed that the 𝘬-means algorithm recognized 4–6 distinctive voltage regions, defining unique periods of the MFC voltage that clearly identify pre- and postinjection and other phases of the MFC lifecycle. This demonstrates that 𝘬-means can identify voltage patterns temporally, which could be further improve the sensing capabilities of MFCs by identifying specific regions of dissimilarity in voltage, indicating changes in the environment.