The United States has thousands of hazardous waste sites, most of which are a legacy of many decades of industrial development, mining, manufacturing and military activities. Subsurface contamination is characterized by considerable spatiotemporal variability and the ensuing health effects are a result of years of exposure. This study focuses on the spatiotemporal representation of the distribution of lead contamination in the groundwater at the Cherry Point Marine Corps Air Station, Havelock, NC. The powerful Bayesian Maximum Entropy (BME) method is used to provide an accurate analysis and determine realistic health effects across space and time. A composite space/time dataset spread out over 14 years of sampling is used. The distribution analysis evidences the systematic decrease in the groundwater lead levels with continuing cleanup operations. The study also analyses the neurological impairment (depression in arithmetic ability in children) and lung cancer effects due to lead. It aims to develop a general exposure and health effect assessment framework that is flexible enough to consider other contaminants of concern at Superfund sites. Finally, the frame work is extended to include demographic information and hence estimate the population impact due to exposure to the contamination. The numerical implementation of this framework relies on the BMElib numerical library.