Matrix methods to analyze long-range transport of air pollutants

Abstract

To assess air quality constraints and impacts of energy activities, models that account for long-range transport processes, as well as for local effects of meteorological dispersion, are required. At the present state of the art of modeling, separate models are used to estimate the detailed, rapidly varying effects of local sources and the long-term average effects of distant sources. Development of the air transport matrix method was undertaken to provide a simpler, faster method of analysis. The method represents results of comprehensive longrange transport models in a simple, easy to use form. The present report is a description of the concept and methodologies used in developing matrices, a preliminary analysis of those matrices and their properties, and a guide to the types of applications they can serve. Matrices have been generated by BNL for transport of sulfur oxide emissions among the 238 Air Quality Control Regions in the conterminous United States, using their AIRSOX model. PNL has used their long-range transport model and a streamlined calculation method to generate matrices for sulfur oxides and for emitted fine particulates. Matrices have been completed for 4 months of meterological data (one in each season) from 1974. BNL further separates matrices according to three categories of sources: utility, industrial, and area sources. They differ in terms of effective stack heights and detailed distribution of source locations within each AQCR. Matrices have also been calculated at the more aggregated levels of state and Federal region boundaries.