Superfund Research Program
A Holistochastic Approach to Human Exposure Assessment
Project Leader: George Christakos
Grant Number: P42ES005948
Funding Period: 1995 - 2006
Project investigators focused on application of the fundamental holistochastic framework (HSF) of human exposure assessment developed in our earlier book (Spatiotemporal Environmental Health Modeling, Christakos, Hristopulos, Kluwer Acad. Publ, Boston, MA). The HSF leads to an integrated study of environmental exposure and the resulting ecological and health impact. This synergy of theoretical and computational sciences (mathematics, physics and computer science) with toxicology, physiology and epidemiology allows one to model, predict and make science-based human exposure decisions and risk assessment.
The researchers are also building a high quality computational library for the BME and S/TRF-n/m methods. The BME method was presented in Modern Spatiotemporal Geostatistics (Christakos, Oxford Univ. Press, 2000; 3rd print, 2001). Project investigators published an additional book (Temporal GIS, Christakos, Bogaert and Serre, Springer-Verlag, 2002), which includes a powerful computational library of spatiotemporal exposure analysis, currently used by researchers in 15 countries.
In addition, Dr. Christakos' team developed applications of HSF to real-world exposure analysis and risk assessment situations. Datasets of particular relevance to Superfund sites were considered, including aqueous arsenic in the U.S. and in Bangladesh, lead contamination at the Cherry Point site (NC), testing a physicoepidemiologic predictability criterion of causal exposure-effect associations in the case of temperature-mortality data (NC), and composite analysis of space/time mortality data (CA). Also, researchers developed a computational approach for incorporating stochastic advection-reaction PDE describing pollutant distribution in a hydrologic system. These case studies amply demonstrate the usefulness of the HSF of human exposure to establish functional relationships between pollutant exposure, burden on the receptor, health effects and population damages and, thus, to generate realistic assessments of health risks.