Solid waste disposal sites (SWDS) – especially landfills – are a significant source of methane. Although having the potential to be captured and used as a fuel, most of the methane formed in SWDS is emitted to the atmosphere. After the United Nations Framework Convention on Climate Change entered into force in 1994 with the final goal of preventing climatic changes, all the countries that have ratified it were asked to estimate and report their greenhouse gas emissions, including methane. In order to support countries in this task, the Intergovernmental Panel on Climate Change (IPCC) has published three sets of guidelines for national inventories, including sets of equations for calculating the quantity of methane formed as biodegradable waste decays. In addition, the Kyoto Protocol has created the Clean Development Mechanism (CDM) to assist the developed countries to offset their own greenhouse gas emissions by assisting other countries to achieve sustainable development and to reduce their emissions. Based on IPCC s methodologies, the CDM has issued a tool to help developers estimate reductions in methane emissions as a result of their project activities. Unfortunately, the four methodologies for calculating methane formation in landfills that are used worldwide – three from IPCC and one from CDM – yield different results, although they are all based on equations to simulate first order decay of biodegradable waste. Furthermore, differences in results from the use of the different models are not clearly presented, and there is not a clear understanding on how they should be used. The incorrect application of the methodologies can be seen in national inventories and in CDM projects activities. The purpose of this thesis is to assess the mathematical models used to predict the generation of biogas by landfills in Brazil and to compare the forecasts with the monitored results over the years of operation. According to Scharff and Jacobs (2006), the emission from a landfill has a high temporal and spatial variability and the authors assert that this is a complicated area of study. Approximately half the volume of this biogas is methane, which is its most significant part; firstly, because it is a greenhouse gas; and secondly, because its burning, as well as being desirable, can generate energy – of the renewable type. For this reason, landfill projects that burn methane are able to receive financial incentives – the carbon credits – through the CDM. Since 2005, when the Kyoto Protocol entered into force and launched the CDM, solid waste disposal sites are seen differently in Brazil. The landfills studied here are CDM projects that had their potential for biogas EXPANDED ABSTRACT evaluated at their inception and which later on had their generation models and their parameters re-evaluated. These projects have undergone a renewal of their first 7-year crediting period, within the CDM procedures, when it was necessary to update the baseline and monitoring methodology. In order to have an ex-ante estimation of the methane generation in CDM landfill projects, it is necessary to follow the methodologies approved by the CDM Executive Board. These methodologies are based on the procedures used for national greenhouse gas inventories of the IPCC, which in turn assesses the scientific knowledge around the world on climate change and greenhouse gases. There are three editions of IPCC guidelines (1997, 2000 and 2006), other than several versions for the ones issued by the CDM Executive Board. This thesis shows the differences among the models contained in these methodologies, besides the different interpretations and different ways to apply the methodologies in the analyzed projects. Two more models were added to the analysis: those referred to in World Bank (2004) – the so-called Scholl-Canyon model and the U.S. EPA s LFG Emissions Model (LandGEM). All CDM projects are required to have a prior design document as well as verified monitoring reports, which demonstrate the emission reductions, all documents published on the Internet. The method of work used in this study was to highlight for each project the generation forecast and the amount of methane captured during several years of operation, with the knowledge that these figures are not a simple translation of carbon credits. CDM procedures require some baseline and leakage discounts and allow more credits whenever electricity is generated from the biogas, and these facts have to be understood and taken into account. Models, parameters, results and efficiencies are then presented and compared. Based on the analyses of the models, the authors also indicate the need for a correction in the tool to calculate emissions from landfills as adopted by the CDM Executive Board. This thesis also shows how the experience of the first 7-year crediting period of each project was used in its second period, and makes a critical analysis of the suitability of the use of these prediction models, as well as a confrontation with other studied models, where the data was adequate. This study has the ultimate intent to contribute to enterprises that are designed to improve sanitary conditions