A MODELING PROCEDURE TO PREDICT CANCER RISKS ASSOCIATED WITH INHALABLE OIL DUST FROM KUWAIT OIL LAKES

A model was constructed to estimate cancer risks associated with PM10- bound polycyclic aromatic hydrocarbons (PAHs) from Kuwait oil lakes. The design of the risk model was based on a conceptual ''chain of even ts'' leading from levels of PAH compounds in oil lakes, erosion of oil dust and input into the atmosphere, to contaminant concentration in a ir, to actual human exposure in residential areas. Uncertainties in th e ''chain of events'' model were addressed using Monte Carlo technique s. To identify the exposure duration of concern [duration beyond which risk becomes unacceptable (i.e. Risk > 10(-6))], four exposure durati ons were tested 10, 20, 40, 70 years. 40 years was identified to be th e exposure duration of concern based on the 95th percentile of the ris k distribution. As a result, the acceptability of risk was specified i n terms of a single constraint: on the 95(th) percentile of the risk d istribution evaluated after 40 years of exposure: 0 < Risk(40 y)(0.95) less than or equal to 10(-6). Based on this constraint, it was estima ted that a removal rate of 217,793.27 m(3)/year to be an adequate acti on for risk management. The northern oil lakes were identified as the lakes of most concern when inhalation exposures are considered.