The Air Quality (AQ) database was constructed using ozone data from the Environmental Protection Agency (EPA) Air Quality System (AQS) Data Mart for the study. For each quarter, we calculated the average daily 8-h maximum concentration for ozone from 18 monitors in the study region. This methodology was selected due to the diurnal nature of ozone exposure and to match the temporal nature of the administrative data from the Texas DSHS. In addition, the 8-h maximum exposure estimate used in this study coincides with the measurement statistic used to calculate the EPA 8-h maximum for non-compliance. The point estimates from the 18 monitors were converted to a continuous surface of ozone measurements using ordinary Kriging implemented in ArcGIS Pro. Multiple surfaces were created for each quarter during the period 2007–2016. Kriging was selected due to its wide use in estimating the ozone concentration between ambient monitoring stations (Diem, 2003; Gorai, Tchounwou & Tuluri, 2016) and its low level of measurement errors compared to inverse distance weighting and data averaging (Joseph et al., 2013). We evaluate the appropriateness of the model by constructing the semi-variogram. We selected the ordinary Kriging method and spherical semi-variogram model as it generalized well to all quarters and appeared to not over fit the surface. The mean and variance of the data were calculated and assumed to be constant over space and time. The data were assumed to be isotropic and stationary.