Quantifying uncertainty in global and sub-global socioeconomic and greenhouse gas emissions futures

Projections of future economic development, energy, emissions and climate involve a wide range of uncertainties. These projections often assume idealized policies. We employ a multi-sector coupled human-natural system model to explore both probabilistic parametric uncertainty and deep uncertainty about climate policy. Scenarios are used to capture deep uncertainties about policy design, including the level of policy stringency, the option for international emissions trading, the coverage of land use change emissions, and the availability of negative emissions technologies (e.g. bioenergy with carbon capture and storage, or BECCS). For each “optimistic” and “pessimistic” combination of policy design assumptions, we sample from probability distributions for model parameters such as total factor productivity growth, population, energy efficiency trends, costs of advanced technologies, fossil fuel resource availability, climate sensitivity, ocean heat uptake and aerosol forcing. We then assess the resulting uncertainty of key outcomes of interest at global and sub-global (regional, sectoral and technology) levels. This uncertainty characterization helps to inform policy discussions and decision-making. We show the impact of policy design assumptions on uncertainty in the distribution of emissions across regions, sectors and greenhouse gases, as well as energy and technology mixes and the cost of the policy. Several insights emerge, such as how failing to cover land use emissions can result in total emissions above the intended cap; how the availability of BECCS and credits for land use emissions can allow for a prolonged use of fossil energy; and how international emissions trading can benefit some regions more than others.