Time series datasets have been used to study the variability in Volatile Organic Compound (VOC) concentration and the effect of atmospheric Oxygen (O2) on their emission from a VOC contaminated site in the US. The datasets were logged by means of an in-borehole gas monitor, the Gasclam (Ionscience, UK). Graphical illustrations of the data show both the VOC concentration and O2 to be temporally variable. However, to determine how much control O2 has on the variability in VOC concentration, Regression Analysis was used. The result shows VOC concentration and O2 to have negative correlations of 0.7134, 0.0058, 0.4558, 0.4105, 0.8162, 0.0982 in boreholes A, C, E, G, I, K respectively. This analysis shows O2 to be the dominant control on the variability in and therefore emission of VOC concentration two (A and I) of the six monitored boreholes in this site. Being the only borehole with the highest negative R2 value, borehole I was reanalysed and the R2 considered over different periods of rising and falling limb O2 with the result showing that O2 was the dominant control on VOC concentrations during 90% of the entire monitoring period. The dominant controls in the variability of VOC concentration in the remaining boreholes need be determined. This is because, understanding their controls will help to predict how VOC concentration will change in future. Also, given that VOC can be dangerous even at trace concentrations; this site is recommended for remediation.