Estimation of methane emission rate changes using age-defined waste in a landfill site

Highlights

• This study proposes a new approach to predict the methane emissions from landfills.

• Age-defined waste samples were collected from a real landfill site.

• Realistic degradation coefficients in the first-order decay model were estimated.

Abstract

Long term methane emissions from landfill sites are often predicted by first-order decay (FOD) models, in which the default coefficients of the methane generation potential and the methane generation rate given by the Intergovernmental Panel on Climate Change (IPCC) are usually used. However, previous studies have demonstrated the large uncertainty in these coefficients because they are derived from a calibration procedure under ideal steady-state conditions, not actual landfill site conditions. In this study, the coefficients in the FOD model were estimated by a new approach to predict more precise long term methane generation by considering region-specific conditions. In the new approach, age-defined waste samples, which had been under the actual landfill site conditions, were collected in Hokkaido, Japan (in cold region), and the time series data on the age-defined waste sample’s methane generation potential was used to estimate the coefficients in the FOD model. The degradation coefficients were 0.050 1/y and 0.062 1/y for paper and food waste, and the methane generation potentials were 214.4 mL/g-wet waste and 126.7 mL/g-wet waste for paper and food waste, respectively. These coefficients were compared with the default coefficients given by the IPCC. Although the degradation coefficient for food waste was smaller than the default value, the other coefficients were within the range of the default coefficients. With these new coefficients to calculate methane generation, the long term methane emissions from the landfill site was estimated at 1.35 × 10⁴ m³-CH₄, which corresponds to approximately 2.53% of the total carbon dioxide emissions in the city (5.34 × 10⁵ t-CO₂/y).

Introduction

The landfilling of biodegradable organic materials, such as kitchen waste, papers and woods causes long term methane gas generation, which can contribute to global warming if not adequately collected by a gas-recovery system. Emissions from landfill sites account for 30% of the total anthropogenic methane emissions in Europe, 34% of those in the US, and 10% of anthropogenic methane emissions worldwide (Perdikea et al., 2008). The EU landfill directive (European Union, 1999) required EU member countries to reduce biodegradable organic waste going to landfill sites and set a goal of reducing the amount of biodegradable waste going to landfills by 65% by 2020 compared to 1995 levels. The evaluation and monitoring of methane emissions from landfill sites is meaningful for preventing global warming.

In general, there are two different approaches to estimate methane emissions from landfill sites: the direct approach, which is based on landfill gas (LFG) emission measurements from the landfill surface, and an indirect calculation based on a straightforward mass balance equation between methane production, recovery and oxidation at the landfill site (Bella et al., 2011).

In the direct approach, methane emissions can be estimated (1) from the soil gas concentration profiles of methane through the landfill cover materials or (2) by using the static and/or dynamic closed flux chamber method, which is applied to a relatively small part of the landfill surface (Bella et al., 2011). However, the direct approaches have limitations, because the measured values are only obtained at discrete points and for a limited short time of the measurement (Spokas et al., 2003). There are spatial and time variations in landfill methane emissions. To reduce the influence of these variations, some larger area measurement assessments, such as a dynamic plume measurement using CO and N₂O as tracer gasses was proposed (Scheutz et al., 2011). However, scaling the direct measurement approach to long term methane emissions from landfill sites is difficult because the measurement is based on the limited temporal period of the measurement.

In contrast, for the indirect calculation, several models have been developed with different orders of kinetics, including zero-, first-, and second-order models, as well as some more complex models (Freidrich and Trois, 2011; Meima et al., 2008). Most of the models are based on the first-order decay (FOD) model to predict methane generation as a surrogate for landfill methane emissions, including the Intergovernmental Panel on Climate Change (IPCC) model (IPCC, 2006), TNO model (Oonk et al., 1994), GasSim Lite model (Golder Associates, 2010), LandGEM model (US-EPA, 2005), and Afvalzorg model (Scharff, 2010). However, the coefficients introduced in the FOD models typically overestimate LFG production because the coefficients are derived from a calibration procedure in ideal steady-state conditions (e.g., gas production factors, conversion coefficients for organic matter degradation) (Cossu et al., 1996). These ideal steady-state conditions differ significantly from actual landfill site conditions. Amini et al. (2012) evaluated the uncertainty in estimated landfill gas generation rates, including the default coefficients given by the IPCC. The uncertainty in the modeled LFG generation rate varied from ±11% to ±17% while the landfills were open, from ±9% to ±18% at the end of waste placement, and from ±16% to ±203% at 50 years after waste placement ended.

Other than the amount and composition of waste, such as the biodegradable organic fraction, there are a variety of factors influencing methane generation, including climate conditions, such as temperature and precipitation, landfilling methods, such as waste compaction, leachate recirculation, and anaerobic or semi-aerobic landfilling (Freidrich and Trois, 2011, Lou and Nair, 2009). Thus, the coefficients in the FOD model should be determined under more realistic landfill site conditions.

This study attempts to establish a new approach to estimate more accurate methane generation rate coefficients through investigation of a real landfill site that has been operated for over 20 years in a city of Hokkaido, Japan. In the new approach, age-defined wastes are sampled from the real landfill site. The time of year when each waste was landfilled is defined based on working records of the landfill. Furthermore, these wastes have been in the landfill under real circumstances for specific periods of time. Although the new approach requires some assumptions and prerequisites in terms of waste composition, the degradation rate, and sample representativeness, the new approach can be an improved method for estimating the amount of long term methane generation from landfill sites.