The scale of the required changes is huge, and time is limited if we are to avoid the most severe effects of climate change. To reduce greenhouse gas emissions from road transport, several fuels and electricity originating from renewable energy sources have been proposed, all of them in different stages of development and with various and shifting environmental impacts. This thesis aims at increasing the usefulness of environmental assessments of emerging technologies as a basis for strategic technology choice. Recommendations for the design and interpretation of such assessments are presented, with a special focus on life cycle assessment (LCA) methodology. A long time perspective, the possibility of system change, and the inclusion of socio–technical change processes allows for the revision of methodological assumptions normally made in LCA of current products. To guide the selection of technologies, there is need for assessment both of technology and of interventions. For the assessment of technology, an attributional approach is applied. Paper I discusses and tests the feasible futures and future performance to be considered in attributional LCAs. The results indicate that the environmental impact attributable to a number of selected fuels, as well as the ranking of them, largely depend on assumptions regarding background systems and by-product use. For the assessment of interventions, a consequential approach is applied. Extensive studies of socio–technical change processes contribute insight into relevant cause–effect chains that can be included in environmental assessments of emerging technologies. A comparison between the Swedish and the Dutch innovation systems for renewable fuels reveals the unfolding of dynamics influenced by shared background factors (Paper II). An investigation of the Swedish history of alternative fuels is used in developing a framework for analysing interaction between emerging technological systems (Paper III). Insights into socio–technical change processes are then used to elaborate scenarios for the future development of renewable fuels in Sweden resulting from current policy choices (Paper IV). In a final paper (Paper V), historical and future cause–effect chains are taken into account in a consequential LCA of ethanol of varying origins in Sweden for the 1990–2020 period. It is concluded that for emerging technologies in an early stage of development, the contribution of an intervention to system change may be more important than the direct change in environmental impact. Finally, it is suggested that all aspects of socio–technical change and the resulting environmental impact may not have to be included in quantitative environmental assessments, such as LCA. ‘Environmental assessment’ could very well include a group of parallel studies that illuminate different cause–effect chains resulting in changed environmental impact, and that are part of a society-wide learning process.