Atmospheric dispersion modeling to assess the tracer dilution method for measuring landfill methane emissions

Friday, 19 December 2014
Diane Taylor1, Madjid Delkash2, Fotini K Chow1 and Paul Thomas Imhoff2, (1)UC Berkeley, Berkeley, CA, United States, (2)University of Delaware,, Newark, DE, United States
Landfill methane emissions are difficult to estimate due to limited observation and uncertainty of the data. The tracer dilution method is a widely used approach that uses a tracer gas released at a known rate from one or more point sources, and the ratio of the concentration of tracer gas to concentration of methane measured at a downwind point is used to calculate the methane emissions rate. Here we use a high-resolution atmospheric model to examine the set-up of the tracer dilution method and its effects on the accuracy of methane emissions calculations. This method relies on optimal weather conditions and is limited by availability of locations where downwind measurements can be taken. Therefore using limited measurements taken with this method to estimate annual landfill emissions will yield totals of dubious accuracy.

The Weather Research and Forecasting model (WRF) is a mesoscale meteorological model that is commonly used for atmospheric research as well as operational forecasts. Here, a scalar tracking subroutine is added to WRF to simulate the methane emissions from the surface of the landfill and the tracer gas from point sources. Using this model, many different tracer release configurations (number and placement of tracer release points and downwind measurement locations) are simulated and compared. Wind speed dependence of methane emissions is examined by prescribing surface flux as a function of local wind speed. The tracer dilution method can only collect landfill emissions data during ideal weather conditions, so modeling emissions during non-ideal conditions will give a better idea of how to predict total annual emissions taking into account the emissions on days when emissions cannot accurately be measured. The WRF output is compared to output of an analogous model adapted from the existing atmospheric model Advanced Regional Prediction System (ARPS) and to observation data from Sandtown Landfill in Delaware, USA. Future work includes adding methane transport to the soil model component of WRF to gain a more accurate understanding of diurnal and seasonal variations in methane emissions.