Evaluation of optimal model parameters for prediction of methane generation from selected U.S. landfills. (15th May 2019)

• Record Type: Journal Article
• Authors: Sun, Wenjie; Wang, Xiaoming; DeCarolis, Joseph F.; Barlaz, Morton A.
• Is Part Of: Waste management. Volume 91(2019)
• Page Start: 120
• Page End: 127
• Publication Date: 2019-05-15
• Subjects: Landfills -- Landfill gas -- Methane -- Gas collection efficiency -- LandGEM; Hazardous wastes -- Periodicals; Refuse and refuse disposal -- Periodicals; 363.728
• DOI: 10.1016/j.wasman.2019.05.004
• Languages: English
• ISSNs: 0956-053X

Abstract

Highlights: Methane collection data from 21 US landfills were used to estimate the best fit k . Inverse modeling was used considering a time-varying gas collection efficiency. The results show the wide variation in the best estimate of k (0.04–0.20) . The result is confounded by the fact that the best fit L0 varied from 70 to 160. Abstract: In practice, methane generation at U.S. landfills is typically predicted by using the EPA's Landfill Gas Emissions Model (LandGEM), which includes two parameters, the methane production potential ( L0, m 3 CH4 Mg −1 wet waste) and the first-order decay rate constant ( k, yr −1 ). Default parameters in LandGEM ( L0 = 100 and k = 0.04) were determined using data that reflect landfill management practices in the 1990s. In this study, methane collection data from 21 U.S. landfills were used to estimate the best fit k by inverse modeling of measured methane collection data in consideration of a time-varying gas collection efficiency. Optimal values of k were identified at a range of L0 s between 55 and 160. The best fit k was greater than the U.S. EPA's default parameter of 0.04 yr −1 at 14 of the 21 landfills studied. Surprisingly, the best fit k was often observed at L0 values greater than 100 m 3 CH4 Mg −1 wet waste which again is the U.S. EPA default. The results show that there is wide variation in the best estimate of k . While there was a tendency for landfills, or sections of landfills that received more moisture to exhibit higher decay rates, the results were not consistent. Some landfills exhibited high decay rates even though the data suggested that they were relatively dry while some wet landfills exhibited low decay rates. The results suggest that L0 captures many factors and that the data may be most useful for site specific analysis as opposed to general landfill predictions.