Characterizing Dominant Processes in Landfills to Quantify the Emission Potential

Our ever-growing amount of solid waste puts a burden on future generations and the environment due to emissions of contaminants such as CO2, CH4, Cl- and heavy-metals for hundreds of years. It is therefore essential that landfill after-care methods are developed that reduce the emission potential of landfills to acceptable levels within the time-span of one generation. Several treatment methods such as aeration and leachate recirculation have shown promising results in reducing concentrations of problematic compounds in leachate and landfill gas emissions. However for application as full-scale technologies, long term evidence of sustainable reduction in emission potential has yet to be provided in practice. It is not possible to measure emission potential directly. Predictions of future emissions from landfills require emission modeling where emission potential is a crucial parameter. The aim of the research presented in this thesis is to present a conceptual modeling approach which increases the confidence in such long term predictions by reducing
the parameter and model uncertainty in a systematic way. As such the approach allows us to quantify the emission potential. Chapter 2 and 3 of this thesis present an approach to develop and select biochemical and physical process networks in a generic conceptual model that allows us to optimally describe measured emissions from lysimeter experiments under anaerobic and aerobic conditions. These networks give a detailed description of the mass balances of contaminants and bacteria in the solid, liquid and gas phase. As a consequence, main emission pathways and rate-limiting processes are identified. Our results give strong indications that only a relatively small amount of the solid waste material present contributes to the measured emissions. The toolbox developed for this thesis, integrates information from different databases with approaches to obtain and couple thermodynamic/kinetic parameters and processes in order to efficiently evaluate a wide variety of networks via Bayesian inference using quantitative criteria. In chapter 4, the optimal biochemical and physical process networks calibrated at the lysimeter and column scale, are applied to predict the emissions at landfill scale. This is achieved by coupling the process networks to a water balance model that calculates the leachate production using a stochastic residence time distribution of water within the waste-body. The parameters of the stochastic residence time model are obtained by optimization using daily leachate production, rainfall and evaporation measurements. After calibration, the decrease in mass of different contaminants present in the waste body, gives a quantitative estimate of the full scale emission potential as a function of time. Results are shown for measured time series of leachate quantity and leachate quality (e.g. Cl–, Na+ and NH4+), but can easily be extended to other parameters. In chapter 5, the effectiveness of different aeration strategies is investigated based on modeled distributions of oxygen throughout a waste-body. The model is based on Darcy’s law for two-phase flow with parameters measured in laboratory experiments. Modeled gas extraction rates are in reasonable agreement with extraction rates measured at landfills. The results present optimal well configurations and aeration strategies for effective treatment. The thesis concludes with a list of the most important research steps for reducing the uncertainty in the approaches for quantification of full scale emission potential in the near future.