Summary In this PhD thesis I studied the influence of biochar discourses on the political practices in Brazil and the impact of biochar on soil organic carbon (SOC) stocks, thus contributing to the current debate on the potential of biochar to mitigate climate change. Biochar is the solid material obtained from the carbonization of biomass. The deliberate production and application to soil distinguishes biochar from other carbonized products, e.g. charcoal. Inspired by the aged charcoal found in the fertile Amazonian Dark Earth (ADE; also known as Terra Preta de Índio), the current application of biochar in soil is claimed to simultaneously address four global challenges: food production, climate change, energy supply and waste reduction (Chapter 1). Biochar is supposed to be an absorbent and stable material, which can be used to retain nutrients in the soil, increasing agricultural productivity, while sequestering carbon over extended periods of time. Therefore, biochar is claimed to be a means to mitigate global climate change. Furthermore, if biochar is produced in a modern pyrolysis plant, it also can co-produce bio-oil and syngas that could be used as energy. And if biochar is produced by carbonization of agricultural residue, biochar may reduce the quantity of solid waste that needs to be disposed of. In Chapter 2, I analysed the policy arrangement related to biochar along the four dimensions of the policy arrangement approach, which are actors, discourse, power and rules. I focused on Brazil, which is an important player in the international biochar debate. My analysis shows that scientists in research institutions are the dominant players in the network, while policymakers, businessmen and farmers are marginally positioned. Experts from Embrapa occupy central positions and thus exercise most power in the network. Moreover, experts linked to ADE have lost prominence in the network. The cause for this reduction was the shift from the ADE/biochar to the biochar/technology discourse. The latter discourse includes different coalitions, such as: ‘climate change mitigation’, ‘improvement of soil fertility’ and ‘improving crop residue management’. Although the biochar/climate coalition is dominant at international level, it is far less prominent in Brazil. Nationally the discourses of ‘improvement of soil fertility’ and ‘improving crop residue management’ have particularly prompted actors’ relationships and practices. However, the biochar/technology discourse is not (yet) institutionalized into formal rules in Brazil. As a consequence, the country lacks an established biochar policy field. Brazilian biochar practices focus on the carbonization of the available residues into biochar and on the application of biochar in soils to increase the SOC content and consequently the fertility of these soils. In this context, in Chapter 3 I tested in the field the potential of biochar produced in traditional kilns to increase the C contents of sandy savannah soils. My results show that biochar produced in traditional kilns is less thermally altered than that produced by industrial kilns and therefore rapidly decomposes. The decomposition rate of traditionally produced biochar was higher (decomposition constant k = 0.32-1.00 year-1) than generally assumed (k = 0.0005-0.005 year-1), and higher than the decomposition of native SOC (k = 0.22 year-1). In Chapter 4 I demonstrated in a short-term laboratory experiment that oilseed-derived biochar had a similar or higher decomposition rate than native SOC. My results show that all three tested oilseed biochars decelerate the decomposition of SOC in the biochar-amended soils, with biochar richer in aromatics having a stronger negative effect than biochar richer in aliphatics. Therefore, oilseed biochar directly increases soil C stocks and indirectly raises soil C sequestration in the short term through decreasing the decomposition of native SOC. In my research, the decomposition studies were performed using 13C isotope analysis. However, the 13C isotope analysis cannot be used when the differences of 13C isotope abundance between biochar and soil are not sufficiently large. Therefore, its use can be limited. In Chapter 5, I aimed at improving the benzene polycarboxylic acid (BPCA) method. I re-designed the protocols of the BPCA method and found a better and faster way to quantify and characterize the BPCAs derived from biochar, compared to the previous protocols. The improved method was then successfully tested and implemented in a laboratory in Brazil. Combining my findings with results of the literature, I conclude (Chapter 6) that there is no evidence that biochar is a reliable way for C sequestration in sandy soils under savannah environments. Biochar decomposition is highly variable, depending on charring conditions, soil and climate: (i) biochar produced by traditional kilns is less thermally degraded than those pyrolysed by industrial kilns; (ii) in sandy soils less biochar accumulates than in clay-silt soils; and (iii) warm-dry conditions raise the decomposition of biochar. These conclusions have a direct consequence for the development of policies on biochar, because we cannot ensure that biochar will sequester the same quantity of C for the same period at different geographical regions.