Skip to content


Quantifying methane emissions from landfills is important for meeting regulatory reporting requirements. Emissions estimates or measurements may also be useful for managing landfill gas collection systems. Traditionally, whole-landfill methane emissions estimates are determined with simple mass balance models using waste properties, landfill gas collection data, landfill cover properties, and estimates of methane oxidation in cover soils. More sophisticated predictive models have also been recently developed, such as CALMIM (Spokas et al., 2011), which include mechanistic models describing methane diffusion and oxidation in landfill covers. Alternatively, whole-landfill methane emissions measurements are possible using downwind gas concentrations in the landfill gas plume. The tracer dilution method (TDM) is one such technique that has received considerable attention in the past five years because of advances in field instrumentation that allow measurements in less time, at smaller cost, and with less expertise than in previous applications. While the TDM is increasingly used to quantify methane emissions, the accuracy of the method has not been firmly established. The key assumption required for accurate measurements is that methane and a tracer gas emitted from the landfill must be well-mixed in the downwind plume sampled. While this assumption has been evaluated in small-scale controlled releases on flat terrain, it has never been evaluated at a real landfill. 

The primary objective of this study was to evaluate the accuracy of the TDM using atmospheric dispersion modeling of wind and emissions from a landfill surface. This is the first application of atmospheric dispersion modeling to landfills that we are aware of. The Sandtown Landfill (Sandtown, DE, USA) was selected in this study, and two atmospheric models were developed for the site that include actual topography and wind conditions, measured with a wind monitor on the landfill surface and compiled from a national weather forecasting program, corresponding to a field measurement campaign using the TDM in 2010. After validating the atmospheric dispersion model for the site, the model was used to evaluate the accuracy of the TDM for different transect distances downwind, transect orientation, vehicle speed (used to sample gas) along the transect, placement of tracer gas release, landfill topography, and wind direction. These simulations show that error in emissions estimates generally decreases with increasing transect distance downwind, is slightly sensitive to transect orientation and vehicle speed, but is most sensitive to the locations of tracer gas release on the landfill surface. Measurement error is minimized by placing the tracers as close to the center of methane emissions as possible, both horizontally and vertically. Measurement errors ranged from -12% (underprediction) to 42% (overprediction), with the largest errors observed with the transect closest to the landfill and when tracers were released upwind of the methane emissions hot spot. While application of the TDM to other landfills will likely yield different measurement errors, the general rule of thumb to release tracers in the center of mass of methane release (both vertically and horizontally) will likely minimize measurement error at other locations. Simulation also examined the role of landfill topography and wind direction, showing that both can influence measurement error as they affect plume dispersion. Thus, the accuracy of the TDM will likely vary with landfill age and thus topography, and may also vary with day of measurement if wind directions vary significantly from day-to-day at a site. 

A secondary objective was to evaluate if observed temporal variability in methane emissions from the TDM field data may be due to temporal variations in emissions at the landfill surface associated with wind, measurement error, or other factors. While changes in barometric pressure may be one such factor, these changes were not significant over the time scale of our field measurements. An atmospheric model was used to explore these questions and to test three hypotheses about the observed correlation between downwind measured methane emissions with the TDM and wind speed on the landfill surface. The modeling disproved two hypotheses, but was not able to disprove the hypothesis that measured methane emissions variation was due to variations in wind speed: emissions increased when wind speed increased in the model, and downwind TDM data reflected this. High wind speed causes short term pressure pulses that may enhance mixing between the landfill cover soil and the atmosphere. While this result is specific to the Sandtown Landfill, it demonstrates that variability in TDM emission measurements may be due to actual emissions variability and not experimental measurement error. If whole-landfill methane emissions measurements vary with wind speed, as indicated in this exercise for data collected from the Sandtown Landfill, then TDM data that are typically collected for particular wind conditions may not be representative of emissions under different atmospheric conditions. 

An increasing number of TDM measurements of methane emissions have been reported at landfills in the US and in Europe. These data are important for guiding regulators and the waste management industry on the importance of greenhouse gas emissions from landfills and the utility of existing approaches to estimate these emissions. In Denmark, recent TDM measurements at 15 landfills indicate that a national greenhouse gas inventory conducted in 2011 likely overestimated landfill emissions by approximately 50% (Mønster et al., 2015). However, results from that study may be challenged, since the accuracy of the TDM for measuring methane emissions from landfills has not been firmly established. This study addresses this problem by applying atmospheric dispersion modeling to quantify TDM measurement error. Mean measurement errors in this study were always less than 42%, sometimes much smaller, and usually resulted in an overestimation of methane emissions. While extrapolating the results from this study at a single landfill to other landfills requires future atmospheric modeling at other sites, this work suggests that TDM measurement errors may usually be less than 40%. This information will be helpful to those using TDM methane emission measurements to develop regulatory policy.