Evaluation of backward Lagrangian stochastic (bLS) model to estimate gas emissions from complex sources based on numerical simulations

Abstract

The focus of the paper is application of an inverse-dispersion technique based on a backward Lagrangian stochastic (bLS) model in order to calculate gas-emission rates from industrial complexes. While the bLS technique is attractive for these types of sources, the bLS calculation must assume a spatial configuration for the source. Therefore, results are presented herein of numerical simulations designed to study the sensitivity of emissions calculations to the assumption of source configuration for complex industrial sources. We discuss how measurement fetch, concentration sensor height, and optical path length influence the accuracy of emission estimation. Through simulations, we identify an improved sensor configuration in order to reduce emission-calculation errors caused by an incorrect source-configuration assumption. It is concluded that, with respect to our defined source, the optimal measurement fetch may be between 200 m and 300 m; also, the ideal measurement height is probably between 2.0 m and 2.5 m. With choices within these two ranges, a path length of about 200 m is adequate, and greater path lengths, above 200 m, result in no substantial improvement in emission calculations.