Analysis of an Inverse Gaussian Plume Modeling Method for Estimating Methane Emissions from Underground Pipeline Leaks

Improving the accuracy and efficiency of methane emission detection is a vital aspect to improving safety and reducing greenhouse gas emissions where leaking natural gas infrastructure is concerned. A mobile technique known as EPA Method 33a was used to estimate methane emission rates from previously surveyed underground leaks in three locations, New York, New Jersey, and Los Angeles in mainly suburban environments. Method 33a requires measurement of ambient methane concentrations for 5 to 15 min downwind of a source along with measured wind and turbulence data. These data are used to estimate the maximum plume concentration and the Gaussian diffusion coefficients for use in the inverse calculation. For the method evaluation, different methods for estimating these parameters were evaluated using measured leak rates or in some cases controlled methane releases. Linear regressions and method difference graphical tests were employed to determine the overall accuracy of Method 33 for each of the different parameter sets used in the inverse method.