The objective of this project is to provide a sound theoretical basis for estimating gaseous emission fluxes at soil/air interfaces which arise from volatilization of subsurface contaminants, including those released at leaking or uncontrolled treatment, storage and disposal sites for hazardous waste. Research on this topic involves the theoretical development of a physically based mathematical model for predicting soil/air fluxes of contaminants, application of geostatistical and other optimal estimators for calibrating model parameters and quantifying model uncertainty at the field scale, and validation of the model using field data. The project emphasizes consideration of the gas phase transport and sorption processes in the unsaturated zone for volatile organic compounds and will explicitly represent temporal and spatial heterogeneities in contaminant and media properties, soil temperature changes, and determined in simulations of representative field conditions. Kriging and Bayesian estimation will be used to calibrate spatially and temporally correlated model parameters. A data collection effort will supplement the limited field data available for model validation. Emissions to the air of hazardous substances released at sites where they are buried may pose significant environmental and health risks, however, the current capability to assess and predict these emissions is extremely limited. Results of this project are likely to provide a better basis then is possible now for engineering design of soil-containment systems for hazardous wastes.