Quantification of Variability and Uncertainty in Stationary Natural Gas-fueled Internal Combustion Engine NOx and Total Organic Compounds Emission Factors

Quantitative methods for characterizing both variability and uncertainty are applied to case studies of emission factors for stationary natural gas-fueled internal combustion engines. NOx and Total Organic Carbon (TOC) emission data sets for lean burn engines were analyzed. Data were available for uncontrolled engines and for engines with pre-combustion chamber (PCC) and "clean burn" NOx control approaches. For each data set, parametric probability distributions were fit to the data using maximum likelihood estimation to represent inter-engine variability in emissions. Bootstrap simulation methods were used to quantify uncertainty in the fitted distribution and uncertainty in the mean emission factor. Some methodological challenges were encountered in analyzing the data. For example, in one instance, only five data points were available, with each data point representing a different market share. Therefore, an approach was developed in which a parametric distribution was fitted to population-weighted data. The range of uncertainty in mean emission factors ranges from approximately plus or minus 10 percent to as much as minus 60 percent to plus 80 percent, depending on the pollutant, control technology, and nature of the available data. The wide range of uncertainty in some emission factors emphasizes the importance of recognizing an accounting for uncertainty in emissions estimates. The skewness in some uncertainty estimates illustrates the importance of using a numerical simulation approach that does not impose restrictive symmetry assumptions on the confidence interval for the mean. In this paper, we briefly present the probabilistic analysis method, the data sets, the results of the analyses, and key findings and recommendations. Recommendations include reporting requirements for emission factor data.